mikeprasad
UserApplied reasoning, persistent agent memory, context engineering, and human-governed trust for AI coding agents — Claude Code, Claude Cowork, Codex, Cursor, Antigravity. Compatible with Obsidian.
Categories
Indexed Skills (31)
foundational-review
Run the foundational review chain on a project BEFORE an irreversible decision (version freeze, format/schema/spec tag, public API or repo flip, major re-scope, or a 'should we keep building this' moment). Verdict-led review → design spec → cold-executable plan → composed /prospect → paste-ready executor kickoff. Optional --extend runs the system-design extension loop. REQUIRES an irreversible-decision anchor — no decision named → redirect to /prospect (plan pre-mortem) or /readiness-audit (surface audit). Triggers: '/foundational-review <scope-root>', '/foundational-review <scope-root> --decision "<the irreversible decision>"', '/foundational-review <scope-root> --extend', 'foundational review', 'review this at the foundations', 'is this the right thing built the right way'. (Code port — ADR-094.)
readiness-audit
Audit a surface against a checklist to answer 'is it clean/legal/consistent to ship for THIS event?' — the recurring sibling of /foundational-review. Parallel exploration per surface, controller re-verification of every load-bearing claim, then tiered evidence-celled findings and a phased remediation plan (findings triage to phases, NOT a shipping list). Read-only probes only. No irreversible-decision anchor required (contrast /foundational-review). Triggers: '/readiness-audit <scope-root> --for "<event>"', '/audit-ready', 'readiness audit', 'release-readiness audit', 'is this ready to ship/publish/hand over', 'public-release audit'. (Code port — ADR-094.)
audit
Umbrella for the audit family. Routes to the sub-audits: knowledge (backlog→knowledge promotion), config (CLAUDE.md/settings drift), style (mine session-log history for revealed working-style rules), and usage (value/ROI report for your own corpus). Use '/audit' for a menu, or '/audit knowledge|config|style|usage|all' to skip it. Trigger: '/audit', '/audit usage', 'run an audit', 'audit everything'.
auto
Drive an autonomous execution arc end-to-end — compose brainstorm→spec→/prospect→plan→/prospect→TDD→/retrospect under the Rule 35 posture, decide objectively-validatable forks yourself, and stop only on a load-bearing fork or an ungranted approval. Modes: `arc` (default), `execute <plan|spec|ticket-id>` (skip ideation), `plan` (stop at a prospected plan, no code), `config` (guided per-run knob picker). Stackable, one word per axis: `full` (authority — all except push) · `attended`|`unattended` (presence) · `continue`|`stop` (duration) · plus `tickets` and `self-restart`. A bare invocation opens the `config` picker instead of guessing a goal. An explicit grant of autonomous latitude that overrides the standing `autonomy` config for the arc and never writes it. Use when the user hands off a goal, plan, ticket, or SESSION.md with latitude to execute WITHOUT per-step approval — 'combined go', 'run overnight', 'just build it', 'do as much as you can'. ENTRY POINT for a multi-step arc, NOT a single concrete change;
clip-thread
NOTE: this skill requires ~~chat or ~~email MCPs, which are typically only connected in Cowork — the Code variant exists for parity but most users will want the Cowork variant. Capture a chat or email thread from a connected MCP to the knowledge intake. Use when user says '/clip-thread', 'clip this thread', 'save this Slack thread', 'capture this email chain', 'archive this conversation'. Pulls thread content from ~~chat (Slack, Teams) or ~~email (Gmail, MS365) MCP, composes a clipping with thread metadata + body, writes to intake/clippings/ for review at next /audit-knowledge. (Code port — ADR-094.)
clip
Save a URL or text snippet to the knowledge intake for later review. Use when user says '/clip', '/save', 'clip this', 'save this link', 'save this snippet', 'capture this URL'. Quick capture without leaving the session — clipped items are reviewed at the next /audit-knowledge run. (Code port — ADR-094.)
extract-doc
NOTE: this skill requires connected ~~docs MCPs (Notion, Google Docs, Confluence, Box, Egnyte), which are typically only present in Cowork — the Code variant exists for parity but most users will want the Cowork variant. Pull insights from a single doc or page (Notion, Google Doc, Confluence, etc.) into the standard intake backlog. Use when user says '/extract-doc', 'extract insights from this doc', 'pull learnings from this page', 'mine this Notion page for knowledge', 'extract from this Confluence'. Differs from /intake doc (which captures one structured doc artifact with reaction) — extract-doc decomposes a doc into multiple intake-backlog entries for audit routing. (Code port — ADR-094.)
ask
Research a question, check existing knowledge first, draft a knowledge doc from the answer, and save directly to the appropriate category. Use when user says '/ask', 'ask about', 'research and save', 'I want to learn about', 'what is the pattern for'. Skips backlogs — the user reviews the answer in real-time before saving. (Code port — ADR-094.)
backlog
View and manage pending backlog items. Use when user says '/backlog', '/backlog insights', '/backlog clear', 'what's pending', 'show backlogs', 'check backlog status'. (Code port — ADR-094.)
setup
Configure aria-knowledge plugin. Creates or validates a knowledge folder, checks dependencies, sets audit cadences, and writes config. Run on first install or after plugin updates. Trigger: '/setup', 'setup aria-knowledge', 'configure knowledge'.
snapshot
Save the current Antigravity transcript to the knowledge intake on demand. Use when user says '/snapshot', 'snapshot the session', 'save this conversation', 'capture the transcript', or 'archive this session'. Same archival output as the canonical pre-compact hook, but explicit because Antigravity has no PreCompact event. Distinct from /extract and /intake.
codemap
Generate or update a feature-organized CODEMAP.md for any codebase. Scans repos, detects frameworks, identifies features, traces full-stack flows (frontend routes → hooks → state → backend views → models → integrations), and produces a navigable reference with a directory table for selective loading. Supports modes: create (full generation), inventory (quick index only), update (incremental refresh), section (rebuild one section). Trigger: '/codemap', '/codemap create', '/codemap inventory', '/codemap update', '/codemap section <name>'.
context
Load relevant knowledge by topic. Queries the tag index and presents matching promoted files for selective loading into context. Use when user says '/context stripe', '/context api pagination', '/context ss', 'load knowledge about...', 'what do we know about...'. (Code port — ADR-094.)
digest
NOTE: this skill requires connected ~~chat / ~~email / ~~project-tracker / ~~docs MCPs, which are typically only present in Cowork — the Code variant exists for parity but most users will want the Cowork variant. Cross-tool rollup of what's pending, what shipped, and what's blocked across chat / email / project tracker / docs. Use when user says '/digest', 'weekly digest', 'cross-tool rollup', 'what's pending across my tools', 'summarize this week', 'standup digest'. Probes all 4 ~~categories and degrades gracefully — produces a digest even with only 1-2 MCPs connected, surfacing which categories were unavailable. (Code port — ADR-094.)
distill
Turn raw task text into a tiered executable spec per TASK.schema.md. Auto-tiers by complexity (micro/standard/full). Optional --group loads CODEMAPs for cited context. Trigger: '/distill', '/distill --group=<tag> "…"'.
extract
Extract uncaptured knowledge from the current conversation before it's lost to compaction. Use after completing a task, before switching context, before large exploratory work (multi-file reads, codebase scans), or when the user signals session end. Trigger: '/extract', 'extract knowledge', 'capture session knowledge'. Also prompt mid-session: 'Task complete — want me to run /extract?' and 'Switching context — want me to run /extract first?'. (Code port — ADR-094.)
handoff
Generate a passoff package so the next reader can pick up cleanly — for future-you in a new session (typically when context is high and you need to restart) or for a coworker (via brief mode). Default + `auto` modes emit a paste-ready next-session opener as the headline artifact, alongside PROGRESS / CLAUDE / memory updates, commit, and /extract; `brief` mode emits an 80-150 word coworker-facing prose brief instead (Slack/email-ready, no file writes); `snap` mode runs like auto but archives the transcript via /snapshot for later extraction instead of running /extract now — use when context is high. Use when handing off — not when finishing for the day with nothing pending (closing out a finished session with no passoff is a different skill). Triggers: '/handoff', '/handoff auto', '/handoff brief', '/handoff snap', 'hand it off', 'handoff and extract', 'context is full, restart this', 'pass off to next session', 'brief a coworker on this', 'wrap and prompt'. (Code port — ADR-094.)
help
Show available aria-knowledge commands. Use when user says '/help', 'aria help', 'what commands are available', 'list commands', 'what can aria do'. (Code port — ADR-094.)
index
Rebuild the knowledge tag index. Scans promoted files, normalizes tags, flags untagged files, suggests freeform-to-known promotions, detects stale files, suggests cross-references, updates project-to-tag mappings, and regenerates index.md. Use when user says '/index', 'rebuild index', 'update index', 'reindex knowledge'. Also called automatically by /audit-knowledge. (Code port — ADR-094.)
intake
Capture knowledge from outside the conversation. A single URL or text snippet is clipped whole to intake/clippings/; files/directories/globs are bulk-scanned into the backlogs; `extract <source>` decomposes a source into backlog entries; `doc <source>` captures a structured 5-section reflection (claims/worth-keeping/contested/action/reaction); `thread <id>` pulls a chat/email thread via MCP. Use when user says '/intake', 'intake from', 'import knowledge from', 'scan this file for knowledge', 'onboard this project', 'clip this', 'save this link', 'save this snippet', 'capture this URL', 'clip this thread', 'save this Slack thread', 'capture this email chain', 'extract insights from this doc', 'mine this Notion page', 'capture a doc', 'log notes on this doc'. Unlike /extract (current conversation), /intake captures external sources; clipped/bulk items are reviewed at the next /audit-knowledge run. (Code port — ADR-094.)
interview
Interview the user to ELICIT knowledge through dialogue, then stage it to the intake/ tree (the elicit-side counterpart to /extract and /intake which HARVEST existing sources). Three modes: '/interview project' (scope a new project/build), '/interview knowledge' (get a topic out of your head into the KB), '/interview deep-dive' (comprehensively extract the rationale behind an existing-but-undocumented system you built — REQUIRES a basis to review). Cadence (one-at-a-time socratic vs research-then-batch-of-questions) is chosen in-session. Use when user says '/interview', 'interview me about X', 'grill me on X', 'deep dive on X', 'scope this project', 'ask me questions about X'. Stages to intake/projects/ or intake/interviews/ for manual review; never auto-promotes. (Code port — ADR-094.)
meeting-notes
NOTE: paste-text fallback works without MCPs, but ~~docs MCPs (Notion, Confluence, Granola) are typically only connected in Cowork — the Cowork variant has better fidelity for MCP-sourced meetings. Fold a meeting transcript or notes into structured intake. Use when user says '/meeting-notes', 'capture meeting notes', 'fold this meeting transcript', 'process this Granola export', 'archive this standup'. Accepts a ~~docs URL (Notion meeting page, Confluence) OR pasted transcript text ��� unique among MCP-consuming skills in offering a paste fallback when no ~~docs MCP is connected. (Code port — ADR-094.)
prospect
Run a structured pre-mortem on a plan or approach BEFORE execution. Per-step risk enforcement, active evidence-sourcing pass (autonomous lookups + targeted user-asks for anything that could become objective), simpler-alternative discipline, plan-formation diagnosis, action verdicts (PROCEED/SHRINK/SPLIT/DEFER/KILL), and a growing failure-mode pattern library. Triggers: '/prospect' (defaults to plan scope), '/prospect plan', '/prospect session', '/prospect todos', '/prospect file <path>', '/prospect linear <id>', '/prospect branch <name>'. Backward-compat flags (--plan, --linear, --branch, --todos, --session) still accepted. (Code port — ADR-094.)
recap
Read-only orientation — render a scannable table of what just happened so you can situate yourself. Modes: '/recap' (this session), '/recap arc' (the last product arc), '/recap commit' (the last commit), '/recap push' (the last push's commits — what I sent up), '/recap pull' (the last pull's changes — what came down to me). Use when user says '/recap', 'catch me up', 'what just happened', 'where am I', 'recap the session', 'recap the last commit/push/pull'. Plain summary only — never validates or judges (that's /retrospect); never writes to disk. (Code port — ADR-094.)
retrospect
Run a structured retrospective on a shipped commit range, release, deployment, PR, commit, or session. Per-fix validation enforcement, active evidence-sourcing pass (autonomous lookups + targeted user-asks for anything that could become objective), simpler-alternative discipline, re-diagnosis, action verdicts, and a growing failure-mode pattern library. Triggers: '/retrospect' (auto-range), '/retrospect commit <hash>', '/retrospect range <ref1>..<ref2>', '/retrospect pr <num>', '/retrospect session', '/retrospect release', '/retrospect deployment'. Backward-compat flags (--range, --pr, --session, --commit) still accepted. (Code port — ADR-094.)
rules
Quick lookup into working rules. Use when user says '/rules', '/rules 22', '/rules dependencies', 'look up rule about...', 'what rule covers...', or references a specific rule number. (Code port — ADR-094.)
stats
Show knowledge base health metrics — file counts, backlog depth, audit status, codemap dates, tag stats, and coverage gaps. Use when user says '/stats', 'knowledge stats', 'how is my knowledge base', 'show stats', 'knowledge health', 'dashboard'. (Code port — ADR-094.)
statusline
Install or remove the aria-knowledge status-line meter — a context-window progress bar plus the rolling 5-hour and 7-day plan-usage percentages, shown at the bottom of the Claude Code CLI. Use when the user says '/statusline', 'add a context meter', 'show context usage in the status line', 'show my 5-hour usage', 'usage meter', 'enable the status line', or 'turn off the status line meter'. Claude Code only (the status line is a Code feature). Triggers: '/statusline', '/statusline on', '/statusline off', '/statusline status'.
stitch
Build or verify cross-repo STITCH.md linking backend + frontends in a product group. Modes: create, verify, diff, section. Uses CODEMAPs as drift source by default. Trigger: '/stitch create <group>', '/stitch diff <group>'.
sync-decisions
NOTE: this skill requires connected ~~docs MCPs with WRITE access (Notion, Confluence, Google Docs), which are typically only present in Cowork — the Code variant exists for parity but most users will want the Cowork variant. Mirror approved decisions from the knowledge folder out to a connected ~~docs MCP (Notion, Confluence, Google Docs). Use when user says '/sync-decisions', 'mirror decisions to Notion', 'push decisions to wiki', 'sync ADRs to Confluence', 'export decisions externally'. WRITE-side skill — embeds Rule 22 advisory preamble per ADR-016 and requires explicit per-write go-gate. Logs each sync to logs/sync-decisions.md. (Code port — ADR-094.)
wrapup
Close out a session cleanly when the work is done — no passoff intended. Reviews session work, updates PROGRESS.md / CLAUDE.md / memory, commits changes, captures session knowledge via /extract, and confirms everything is wrapped up, captured, and documented. Use when the task is complete and nothing needs to carry into a next session — not for passing work off to a future session or coworker. Triggers: '/wrapup', '/wrapup auto', '/wrapup snap', 'wrap up', 'wrap it up', "I'm done", 'close out', 'finish session', 'end session', 'saying goodbye'. Auto mode applies implicit-yes on all gates and runs silently. Snap mode runs like auto but archives the transcript via /snapshot for later extraction instead of running /extract now — use when context is high. (Code port — ADR-094.)
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.