aiming-lab · rohan-tessl · Mar 23, 2026
diff --git a/memory_data/skills/agent-task-handoff/SKILL.md b/memory_data/skills/agent-task-handoff/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: agent-task-handoff
-description: Use this skill when delegating a subtask to a sub-agent, spawning a parallel worker, or handing off work across sessions. Write a self-contained task description so the receiving agent needs no prior context.
-category: agentic
+description: "Use this skill when delegating work to a sub-agent, spawning parallel workers, or resuming a task in a new session. Helps you write self-contained handoff packages covering goal, context, constraints, inputs, and success criteria so the receiving agent needs zero prior knowledge."
 ---
 
 # Agent Task Handoff

diff --git a/memory_data/skills/async-communication-etiquette/SKILL.md b/memory_data/skills/async-communication-etiquette/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: async-communication-etiquette
-description: Use this skill when writing messages in async channels (Slack, GitHub issues, email threads) where the reader may not have context and cannot ask follow-up questions immediately.
-category: communication
+description: "Use this skill when drafting Slack messages, GitHub issues, pull request comments, or email threads where the reader may not have context and cannot ask follow-up questions immediately. Structures messages with context, a clear ask, options considered, and a deadline so no round-trips are needed."
 ---
 
 # Async Communication Etiquette

diff --git a/memory_data/skills/audience-aware-communication/SKILL.md b/memory_data/skills/audience-aware-communication/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: audience-aware-communication
-description: Use this skill when writing any explanation, documentation, or response that will be read by someone else. Match vocabulary, depth, and format to the audience's expertise level before writing.
-category: general
+description: "Use this skill when writing docs, explanations, READMEs, or any response targeted at a specific reader. Guides you to match vocabulary, depth, and format to the audience's expertise level — beginner, expert, or mixed — before writing."
 ---
 
 # Audience-Aware Communication

diff --git a/memory_data/skills/auth-and-authorization-patterns/SKILL.md b/memory_data/skills/auth-and-authorization-patterns/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: auth-and-authorization-patterns
-description: Use this skill when implementing authentication (login, token issuance) or authorization (access control, permissions). Apply whenever the task involves login flows, JWT, OAuth2, session management, or RBAC.
-category: security
+description: "Use this skill when implementing login flows, JWT token issuance, OAuth2/OIDC integrations, session management, or role-based access control (RBAC). Covers password hashing, token lifetimes, PKCE, claim validation, and least-privilege enforcement."
 ---
 
 # Auth & Authorization Patterns

diff --git a/memory_data/skills/avoid-acting-on-assumptions/SKILL.md b/memory_data/skills/avoid-acting-on-assumptions/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: avoid-acting-on-assumptions
-description: Common mistake — proceeding with assumptions about ambiguous requirements instead of asking a clarifying question first. This skill reminds you to stop and ask before acting on uncertain interpretations.
-category: common_mistakes
+description: "Use this skill when a request leaves the target language, framework, file behavior, or environment unstated and you are about to guess. Prompts you to pause and ask one clarifying question rather than proceeding with an assumption that could produce the wrong result."
 ---
 
 # Avoid Acting on Assumptions

diff --git a/memory_data/skills/avoid-hallucinating-specifics/SKILL.md b/memory_data/skills/avoid-hallucinating-specifics/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: avoid-hallucinating-specifics
-description: Common mistake — stating specific facts (API endpoints, library versions, config options, function signatures) with false confidence when uncertain. Always flag uncertainty rather than guessing specifics.
-category: common_mistakes
+description: "Use this skill when you are about to state a specific API endpoint, library version, function signature, file path, or configuration value that you are not certain about. Prevents confident hallucination by requiring explicit uncertainty flags and placeholder values instead of guesses."
 ---
 
 # Avoid Hallucinating Specifics

diff --git a/memory_data/skills/avoid-scope-creep/SKILL.md b/memory_data/skills/avoid-scope-creep/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: avoid-scope-creep
-description: Common mistake — doing unrequested work (refactoring, adding extra features, cleaning up style) when the user asked for a specific, targeted change. Only change what was explicitly asked.
-category: common_mistakes
+description: "Use this skill when completing a targeted bug fix, small edit, or focused change where you notice other issues and feel the urge to also refactor, rename, or clean up. Enforces the rule of changing only what was explicitly requested and surfacing other issues as notes instead."
 ---
 
 # Avoid Scope Creep

diff --git a/memory_data/skills/clarify-ambiguous-requests/SKILL.md b/memory_data/skills/clarify-ambiguous-requests/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: clarify-ambiguous-requests
-description: Use this skill when the user's request is ambiguous, under-specified, or could be interpreted in multiple ways. If proceeding with a wrong assumption would waste significant work, always ask exactly one focused clarifying question before doing anything.
-category: general
+description: "Use this skill when a request is vague ('make it better'), under-specified (missing file path, model name, or target), or could be interpreted in multiple conflicting ways. Ask exactly one focused clarifying question — not several — before taking any action."
 ---
 
 # Clarify Ambiguous Requests

diff --git a/memory_data/skills/codebase-navigation/SKILL.md b/memory_data/skills/codebase-navigation/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: codebase-navigation
-description: Use this skill when exploring an unfamiliar codebase, tracing code paths, or answering questions about how the system works. Read before writing, and build a mental model of the architecture before making changes.
-category: coding
+description: "Use this skill when exploring an unfamiliar codebase, tracing how a feature works end-to-end, or finding where a function is defined versus called. Provides a structured read-first strategy — entry points, imports, data model, tests, git log — before making any changes."
 ---
 
 # Codebase Navigation

diff --git a/memory_data/skills/context-window-management/SKILL.md b/memory_data/skills/context-window-management/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: context-window-management
-description: Use this skill in long conversations or multi-turn agentic sessions where context may be lost or the conversation is approaching token limits. Summarize, prioritize, and compact context proactively before it becomes a problem.
-category: agentic
+description: "Use this skill when a conversation is getting long, a session is approaching token limits, or key decisions risk being lost in a multi-turn agentic task. Proactively summarize state, prioritize active context, and offload resolved information to file-based memory to keep the working context focused."
 ---
 
 # Context Window Management

diff --git a/memory_data/skills/data-validation-first/SKILL.md b/memory_data/skills/data-validation-first/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: data-validation-first
-description: Use this skill before any data analysis, transformation, or modeling. Always inspect and validate the data before drawing conclusions or writing transformations.
-category: data_analysis
+description: "Use this skill when starting any data analysis, ETL pipeline, or machine learning task. Before writing transformations or drawing conclusions, inspect shape, types, nulls, and duplicates to surface hidden data quality issues early."
 ---
 
 # Data Validation First

diff --git a/memory_data/skills/debug-systematically/SKILL.md b/memory_data/skills/debug-systematically/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: debug-systematically
-description: Use this skill when diagnosing a bug, unexpected behavior, test failure, or any situation where code does not behave as expected. Follow a structured debugging process instead of randomly changing code.
-category: coding
+description: "Use this skill when a test is failing, code behaves unexpectedly, or there is a bug to diagnose. Follow a structured reproduce-isolate-hypothesize-test-fix-verify cycle instead of randomly changing code or guessing at the cause."
 ---
 
 # Debug Systematically

diff --git a/memory_data/skills/do-not-retry-without-diagnosis/SKILL.md b/memory_data/skills/do-not-retry-without-diagnosis/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: do-not-retry-without-diagnosis
-description: Common mistake — retrying the same failing command or API call without understanding why it failed. Always diagnose the root cause before retrying anything.
-category: common_mistakes
+description: "Use this skill when a command, tool call, or API request fails. Stop before retrying — read the error message, classify it as transient or permanent, and fix the root cause first. Prevents wasted retries on errors that will never self-resolve."
 ---
 
 # Do Not Retry Without Diagnosis

diff --git a/memory_data/skills/git-workflow/SKILL.md b/memory_data/skills/git-workflow/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: git-workflow
-description: Use this skill when working with git — making commits, creating branches, resolving merge conflicts, opening pull requests, or reviewing diffs. Apply whenever the user asks about version control operations.
-category: coding
+description: "Use this skill when making commits, creating branches, opening pull requests, resolving merge conflicts, or reviewing diffs. Enforces imperative commit messages, focused single-purpose PRs, and safe staging practices — never force-pushing to shared branches or skipping hooks."
 ---
 
 # Git Workflow Best Practices

diff --git a/memory_data/skills/graceful-error-recovery/SKILL.md b/memory_data/skills/graceful-error-recovery/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: graceful-error-recovery
-description: Use this skill when a tool call, command, or API request fails. Diagnose the root cause systematically before retrying or changing approach. Do not retry the same failing call without first understanding why it failed.
-category: general
+description: "Use this skill when a tool call, shell command, or API request returns an error or unexpected result. Read the full error message and stack trace, identify the root cause (typo, missing dependency, permission, network, logic bug), fix it, then retry — never swallow errors silently or loop on the same failing call."
 ---
 
 # Graceful Error Recovery

diff --git a/memory_data/skills/idempotent-script-design/SKILL.md b/memory_data/skills/idempotent-script-design/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: idempotent-script-design
-description: Use this skill when writing scripts, cron jobs, data pipelines, or any automated process that may be run multiple times. Design every operation to be safely re-runnable without side effects.
-category: automation
+description: "Use this skill when writing scripts, cron jobs, data pipelines, or database migrations that may run more than once. Apply check-before-create guards, upsert patterns, and atomic writes so that re-running the script never causes duplicate data, partial writes, or crashes."
 ---
 
 # Idempotent Script Design

diff --git a/memory_data/skills/input-validation-and-sanitization/SKILL.md b/memory_data/skills/input-validation-and-sanitization/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: input-validation-and-sanitization
-description: Use this skill when implementing any endpoint, form handler, CLI tool, or function that accepts external input. Validate and sanitize all untrusted data before processing — never assume input is safe.
-category: security
+description: "Use this skill when building API endpoints, form handlers, CLI tools, or any function that accepts user-supplied data. Validate type, range, length, and format at the system boundary; prevent SQL injection with parameterized queries; prevent XSS with HTML escaping; prevent path traversal with canonical path checks."
 ---
 
 # Input Validation and Sanitization

diff --git a/memory_data/skills/plan-before-multi-step-execution/SKILL.md b/memory_data/skills/plan-before-multi-step-execution/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: plan-before-multi-step-execution
-description: Use this skill before executing a sequence of 3 or more steps, especially when steps are irreversible or depend on each other. Write out the plan and verify it before starting execution.
-category: agentic
+description: "Use this skill before executing a sequence of 3 or more steps, especially when steps are irreversible, depend on each other, or could leave state in a broken intermediate condition. Decompose into ordered steps, flag irreversible actions for confirmation, and verify each step's output before proceeding to the next."
 ---
 
 # Plan Before Multi-Step Execution

diff --git a/memory_data/skills/prioritize-by-impact/SKILL.md b/memory_data/skills/prioritize-by-impact/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: prioritize-by-impact
-description: Use this skill when the user has a list of tasks and needs help deciding what to do first. Rank by impact and urgency — not order of mention — and surface the highest-leverage work.
-category: productivity
+description: "Use this skill when the user has a list of tasks, items, features, or bugs and needs help deciding what to tackle first. Applies impact-urgency ranking to surface the highest-leverage work — use when someone says 'what should I work on', 'help me prioritize', or 'rank these tasks'."
 ---
 
 # Prioritize by Impact

diff --git a/memory_data/skills/professional-email-drafting/SKILL.md b/memory_data/skills/professional-email-drafting/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: professional-email-drafting
-description: Use this skill when drafting emails, Slack messages, announcements, or any external/internal communication. Apply professional structure and appropriate tone before writing any message.
-category: communication
+description: "Use this skill when writing or improving a business email, Slack message, announcement, or internal memo. Use when someone says 'draft an email', 'write a message to', 'help me respond to', or 'how should I word this'. Produces structured, appropriately toned communication with a clear subject, body, and call to action."
 ---
 
 # Professional Email Drafting

diff --git a/memory_data/skills/robust-error-handling-in-scripts/SKILL.md b/memory_data/skills/robust-error-handling-in-scripts/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: robust-error-handling-in-scripts
-description: Use this skill when writing shell scripts, Python automation, or any unattended batch job. Ensure failures are detected, logged, and handled — never silently ignored.
-category: automation
+description: "Use this skill when writing or reviewing shell scripts, Python automation scripts, cron jobs, or any unattended batch process. Apply when someone asks to 'add error handling', 'make this script fail safely', or 'handle failures in my automation'. Ensures failures are detected with set -euo pipefail, logged with tracebacks, and retried where appropriate."
 ---
 
 # Robust Error Handling in Scripts

diff --git a/memory_data/skills/secrets-management/SKILL.md b/memory_data/skills/secrets-management/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: secrets-management
-description: Use this skill when handling API keys, passwords, tokens, private keys, or any sensitive credential. Never hardcode secrets in source code — apply this whenever the word "key", "token", "password", or "secret" appears in the task.
-category: security
+description: "Use this skill when handling API keys, passwords, tokens, private keys, or any sensitive credential. Trigger on the words 'key', 'token', 'password', 'secret', 'credential', or 'env var' in the task. Prevents hardcoding secrets, configures .gitignore, sets up environment variable loading, and recommends secrets managers for production."
 ---
 
 # Secrets Management

diff --git a/memory_data/skills/secure-code-review/SKILL.md b/memory_data/skills/secure-code-review/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: secure-code-review
-description: Use this skill when reviewing or writing code that handles user input, authentication, file I/O, network requests, or database queries. Always check for common security vulnerabilities before considering the code complete.
-category: coding
+description: "Use this skill when reviewing or auditing code for security issues — especially code that handles user input, authentication, SQL queries, file I/O, or HTTP requests. Use when someone asks to 'review this for security', 'check for vulnerabilities', or 'is this code safe'. Checks for SQL injection, XSS, hardcoded secrets, missing auth checks, and insecure dependencies."
 ---
 
 # Secure Code Review Checklist

diff --git a/memory_data/skills/source-evaluation/SKILL.md b/memory_data/skills/source-evaluation/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: source-evaluation
-description: Use this skill when presenting information from external sources, citing research, or answering factual questions. Assess source credibility and recency before relying on it.
-category: research
+description: "Use this skill when assessing the credibility or reliability of an external source, citation, article, study, or dataset. Use when someone asks 'is this source reliable', 'can I trust this article', or 'how good is this reference'. Applies the CRAAP criteria (currency, relevance, authority, accuracy, purpose) and flags red flags like missing citations or anonymous authorship."
 ---
 
 # Source Evaluation

diff --git a/memory_data/skills/sql-best-practices/SKILL.md b/memory_data/skills/sql-best-practices/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: sql-best-practices
-description: Use this skill when writing SQL queries — selects, joins, aggregations, window functions, or schema modifications. Apply whenever SQL is needed to ensure correctness, safety, and performance.
-category: data_analysis
+description: "Use this skill when writing or reviewing SQL queries including SELECT statements, JOINs, aggregations, window functions, or DDL/DML like CREATE, UPDATE, DELETE. Use when someone asks to 'write a query', 'optimize this SQL', or 'is this query safe'. Enforces column-explicit selects, CTE usage, parameterized queries, transaction wrapping for destructive ops, and EXPLAIN-driven optimization."
 ---
 
 # SQL Best Practices

diff --git a/memory_data/skills/structured-logging-and-observability/SKILL.md b/memory_data/skills/structured-logging-and-observability/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: structured-logging-and-observability
-description: Use this skill when building production services, pipelines, or automation that needs to be debugged, monitored, or audited. Add structured logs, metrics, and health checks before shipping any service.
-category: automation
+description: "Use this skill when adding logging, monitoring, or observability to a production service, API, pipeline, or background worker. Use when someone asks to 'add logging', 'set up monitoring', 'make this debuggable', or 'add a health check'. Configures structured JSON logs with correct log levels, exposes key metrics (rate/error/latency), and adds a /health endpoint."
 ---
 
 # Structured Logging and Observability

diff --git a/memory_data/skills/structured-progress-update/SKILL.md b/memory_data/skills/structured-progress-update/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: structured-progress-update
-description: Use this skill when summarizing progress on an ongoing project or multi-step task. Give a clear, scannable status report whenever asked for an update or at the end of a work session.
-category: productivity
+description: "Use this skill when reporting on the status of an ongoing project, sprint, or multi-step task. Use when someone says 'give me an update', 'what's the status', 'summarize progress', or 'end of day report'. Produces a scannable Done/In-progress/Blocked/Next format that surfaces blockers proactively."
 ---
 
 # Structured Progress Update

diff --git a/memory_data/skills/structured-research-workflow/SKILL.md b/memory_data/skills/structured-research-workflow/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: structured-research-workflow
-description: Use this skill when conducting research on a topic from scratch — literature review, competitive analysis, technical due diligence, or fact-finding. Apply before starting any open-ended research task.
-category: research
+description: "Use this skill when asked to research a topic, do a literature review, competitive analysis, technical due diligence, or fact-finding. Provides a scoped, step-by-step process to search broadly, filter sources ruthlessly, synthesize findings by theme, and flag uncertainty — preventing shallow first-result reporting."
 ---
 
 # Structured Research Workflow

diff --git a/memory_data/skills/structured-step-by-step-reasoning/SKILL.md b/memory_data/skills/structured-step-by-step-reasoning/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: structured-step-by-step-reasoning
-description: Use this skill for any problem that involves multiple steps, tradeoffs, or non-trivial logic. Think out loud before answering to improve accuracy and transparency. Apply whenever the answer is not immediately obvious.
-category: general
+description: "Use this skill when solving math problems, logic puzzles, multi-step planning, debugging, or architecture decisions where the answer isn't immediately obvious. Guides reasoning out loud — restate the question, identify sub-problems, work through each in order, check consistency, then summarize — improving accuracy and transparency."
 ---
 
 # Structured Step-by-Step Reasoning

diff --git a/memory_data/skills/task-decomposition/SKILL.md b/memory_data/skills/task-decomposition/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: task-decomposition
-description: Use this skill when a user presents a large, vague goal. Break it into concrete, ordered sub-tasks before starting any work. Apply whenever the request is larger than a single focused action.
-category: productivity
+description: "Use this skill when a user presents a large or vague goal like 'build this feature', 'set up this system', or 'migrate this codebase'. Breaks the goal into concrete, ordered sub-tasks, identifies dependencies and unknowns, and confirms the plan before execution — preventing wasted effort from jumping in too early."
 ---
 
 # Task Decomposition

diff --git a/memory_data/skills/technical-writing-clarity/SKILL.md b/memory_data/skills/technical-writing-clarity/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: technical-writing-clarity
-description: Use this skill when writing documentation, READMEs, technical specs, runbooks, or any text that explains a system or process to other engineers. Apply before writing any developer-facing document.
-category: communication
+description: "Use this skill when writing a README, runbook, technical spec, ADR, or any developer-facing document. Applies principles of clarity — leading with purpose, active voice, one idea per paragraph, concrete examples over abstract descriptions — and provides a standard runbook/how-to structure with overview, prerequisites, numbered steps, verification, and troubleshooting."
 ---
 
 # Technical Writing for Clarity