agent-spec-builder
Build a Prompt Hardener agent_spec.yaml from an existing codebase (from-code) or through an interactive interview (from-questions). Use when the user wants to create, generate, or scaffold an agent spec, or when they mention agent_spec.yaml creation. Generates agent_spec.yaml, evidence.md, and open_questions.md with confidence tracking and evidence trails.
codex-implementation-loop
Delegate implementation work to Codex (via the codex-companion runtime), then review its diff, send it back to iterate, gate on the full test suite, and ship it as a PR. Use this whenever the user wants Codex to write code, mentions handing off / delegating implementation to Codex, asks to work through a plan or spec unit-by-unit with Codex doing the coding, or wants a review-and-merge loop wrapped around Codex output â and also when resuming such a loop ("keep going", "next unit", "继ç»ä¸ä¸ä¸ª"). It encodes constraints that are expensive to rediscover: Codex runs in the real environment with effectively read-only git, must not be pointed at a full test suite by default, accepts only specific --effort values, and its self-report is a claim rather than evidence.
agents-md-mapper
This skill should be used when creating, refreshing, or validating a repository `AGENTS.md` so it stays concise, current, and grounded in repository evidence. Use when `AGENTS.md` is missing or stale, after refactors or tooling changes, when new docs become the system of record, or when adding lightweight drift checks.
Challenge â `/challenge [gate1|gate2|<file>]`
The cheapest place to lose a gate is in rehearsal. Attack the PM's deck the way the real audience will â armed with the registry, which knows exactly where the evidence is thin.
bug-investigator
Executes diagnostic investigation tasks to test debugging hypotheses. Runs tests, traces execution, checks git history, and reports evidence. (converted from agent)
Brag Sheet â Work Impact Writer
Turn engineering work into evidence-backed impact statements for performance reviews, self-reviews, promotion packets, and weekly updates. Uniquely mines Copilot CLI session logs, git history, and PRs to reconstruct forgotten work.
accessibility-guide
Use when planning, assessing, governing, or operating accessibility across web, mobile, desktop, documents, content, and service journeys. Triggers on WCAG version or level selection, applicable criteria, inclusive design, keyboard or assistive-technology evaluation, accessibility evidence, release gates, conformance claims, exceptions, remediation, monitoring, or exact-revision reassessment, even when the user doesn't say 'accessibility assurance'.
Governed execution layer for AI coding assistants: clarify intent, route capabilities, review evidence, verify results, and write back lessons across Claude Code, Codex, OpenClaw, and Cursor.
Save current project task progress to local task state file for cross-session continuity. Use when the user asks to save progress, uses an equivalent localized trigger phrase, the session is interrupted, or the user wants to resume work later. Writes to .claude/project-task-state.json so next session can load it via SessionStart hook.
Agentic Orchestration (Experimental Working-Model Protocol)
> **Status â experimental, instrumented, validation in progress.** This > skill is shipped to begin collecting evidence about whether a structured > verifiability-router protocol changes Claude's behavior in ways that > measurably help users. `memesh patterns` exposes a local counter so you > can
ac-verify
Lightweight single-pass AC verification before Archive â run implementation against each AC, produce pass/fail evidence. FAIL blocks Archive and rolls back to Phase 4. Mutually exclusive with ultraqa: choose ac-verify when AC count ⤠3 AND risk â HIGH; otherwise use ultraqa. See .claude/rules/skill-precedence.md Zone C.
Opentargets
Open Targets MCP server for targets, diseases, drugs, variants, and evidence
EU Audit Trail
Tamper-evident audit trail MCP server for EU AI Act & GDPR compliance.
Governance layer for Claude Code Agent Teams — durable auditability, operational controls, and evidence trails for AI-assisted development.
Governance framework for AI agent team coordination, audit trails, and boundary enforcement.
context-modes
Structured work modes for agent sessions. Set LACP_CONTEXT_MODE to activate: tdd (red-green-refactor), debugging (4-phase root cause), sprint (pre-agreed criteria), verification (evidence-before-claims), brainstorm (design first), think (pause-and-reflect), orchestrate (task decomposition). Each mode injects behavioral rules at session start.
fact-check
Verify a specific claim by searching for evidence across web and academic sources. Use when the user asks to verify, fact-check, or confirm a statement.
BrowseAI Dev
Evidence-backed web research for AI agents with citations and confidence scores.
curator-evolution
Use when interpreting Hermes Curator Evolver evidence reports, historical session backfill, proposals, verifier results, candidate search, auto-run output, timer installs, or guarded apply manifests.
agent-plan-to-workers
Compile a frozen reviewed plan into deterministic worker packets with dependency waves, ownership, validation, evidence, and result contracts.
agentic-legibility
Score a repository's agentic legibility from repo-visible evidence only. Use when Codex needs to audit how easy a codebase is for coding agents to discover, bootstrap, validate, and navigate, especially for harness-engineering reviews, developer-experience audits, repo cleanup, or before/after comparisons after improving docs, tooling, or architectural constraints.
acquiring-disk-image-with-dd-and-dcfldd
Create forensically sound bit-for-bit disk images using dd and dcfldd while preserving evidence integrity through hash verification.
agent-receipt
Create a compact, evidence-backed proof-of-work receipt for a completed AI agent run. Use when the user wants to summarize what an agent actually built, how long it took, what changed, tests run, interventions, deployment status, or skills used.
hyperfocus
ADHD-friendly output formatting for Codex. Restructures responses with evidence-based cognitive accessibility: chunking, visual hierarchy, front-loaded key points, and progressive disclosure. Three modes: clean, flow (default), zen. Use when user says "hyperfocus", "focus mode", "adhd mode", "adhd friendly", or invokes /hyperfocus.
codex-upcoming-features
Fetch and summarize upcoming unreleased Codex features using a durable local clone synced from GitHub, with source-file mining as primary evidence. Use when asked for latest upcoming/openai-codex features, what is coming next but not in the latest stable release, or a live release-gap summary with links and as-of timestamp.
assessment-design
Evidence-based assessment design with rubrics, feedback strategies, and formative checkpoints. Aligns each assessment to learning objectives using Bloom's taxonomy. Applies Nicol's 7 principles of good feedback practice. Reads from /learning-objectives manifest and extends it with assessment specs. (idstack)
agentic_search
Run a local-first research workflow by planning sub-queries, iterating retrieval, validating sources, and producing traceable evidence-backed conclusions.
specrail-diagnose-ci
Use when diagnosing or fixing CI failures in a SpecRail-governed repository. Collects fresh CI evidence, reproduces failures locally when possible, identifies root cause before fixing, and reports verification without claiming green CI from stale or missing data.
figma-design-to-code
Implement or update project-consistent UI code from a Figma selection or nodeId using TemPad Dev MCP. Use when the user wants visible Figma UI recreated, ported, or integrated into the target project's framework, styling system, tokens, and existing components when available. Do not use for design critique, product invention, generic code review, or for guessing hidden states, responsiveness, or behavior not shown in design or project evidence.
migliora
Improve a project's development workflow from observed failures, repeated work and avoidable waits. Apply scoped, verified remedies and useful project instructions, then automatically commit owned changes when the project's required checks and reviews are complete. Use after recurring development friction or when asked to improve the process; do not add rules without evidence.
code-change-verification
Verify code changes by identifying correctness, regression, security, and performance risks from diffs or patches, then produce prioritized findings with file/line evidence and concrete fixes. Use when reviewing commits, PRs, and merged patches before/after release.
code-remediate
Collect and resolve selected findings in one pass, then verify; bare PR targets use online items, while PR +review authorizes matching review evidence and any needed review step.
Pretorin Compliance
Access Pretorin compliance systems, controls, evidence, and narratives from your AI tools.
evidence-heavy-evaluator
Generate an evidence-first, read-only repository evaluation report with deterministic scoring and actionable recommendations. Use when the user asks to assess readiness, maintainability, release-readiness, documentation gaps, or engineering health and wants auditable artifacts (`json` + `markdown` + raw command logs).
analyze-user
Use when durable cross-project user preferences must be inferred from coding, writing, or analysis evidence and stored as a profile. Not for one-project context recovery, casual preference acknowledgment, or ordinary task execution.
horizon-app-testing
Run or inspect declared iOS and Android native app tests through Horizon device_test_run and app_* MCP tools. Use for App Automate sessions, native actions, evidence, and cleanup; desktop fixtures use horizon-device.
content-scoring
Score content against the 10 GEO criteria with evidence and prioritized fixes. Use when users ask to score, rate, evaluate, or estimate ranking strength.
change-verification
Use after code or configuration changes to prove the changed behavior with minimal sufficient commands, HTTP/browser checks, or static evidence. DO NOT USE FOR: purely speculative review without a concrete change.
browse
Fast headless browser for QA testing and site dogfooding. Navigate any URL, interact with elements, verify page state, diff before/after actions, take annotated screenshots, check responsive layouts, test forms and uploads, handle dialogs, and assert element states. ~100ms per command. Use when you need to test a feature, verify a deployment, dogfood a user flow, or file a bug with evidence. Use when asked to "open in browser", "test the site", "take a screenshot", or "dogfood this".
å¤æ ¼å¼æ件深度ç解ï¼file-readingï¼ââ 常驻横å
ä½ æ¯ Light æè½å ç**æ件ç解å½å±æ¹**ï¼ä»»ä½ä»»å¡ä¸æ¦æ¶å"ç¨æ·ç»çæ件 / å·²æææ"ï¼ä½ åå°èªå¨å¯ç¨ï¼ æå®**读æ**å交ç»ä¸æ¸¸ã头é¨å类已ç»è½åç»ææ½åã论æ深读æ claimâevidence åæï¼ä¸è½æå®ä»¬ç»ç§°ä¸º "åªä¼æ½å"ãLig
ADD â the orchestration engine
You are the orchestrator. ADD keeps the AI fast *and* safe by fixing direction (spec, scenarios, contract, failing tests) **before** the build, and trusting the result through passing evidence rather than a plausible-looking diff.
historic_sql_patterns
Identify recurring cross-table historic-SQL analytical intents from a bounded pattern shard and emit typed pattern evidence for deterministic wiki projection.
biomcp
Search and retrieve biomedical data - genes, variants, clinical trials, articles, drugs, diseases, pathways, proteins, adverse events, pharmacogenomics, and phenotype-disease matching. Use for gene function, variant pathogenicity, trials, drug safety, pathway context, disease workups, and literature evidence.
ieee-acm-paper-writing
Draft, rewrite, compress, structure, calibrate, humanize, or audit engineering manuscripts for IEEE and ACM Transactions, journals, and conferences, with optional self-contained HTML audit maps and script-checked claim ledgers. Use for abstracts, introductions, related work, system models, mathematical formulations, algorithms, experimental methods, results, discussions, conclusions, contribution lists, notation and claim audits, evidence-grounded peer-review reports, venue adaptation, machine-idiom humanization of AI-flavored prose, or landmark-paper calibration in communications, signal processing, energy systems, robotics, optimization, simulation, ML-assisted engineering, and computer systems. Do not use for general documentation, grant proposals, marketing copy, or editorial accept/reject advocacy detached from manuscript evidence.
buyer-eval
Structured B2B software vendor evaluation for buyers. Researches your company, asks domain-expert questions, engages vendor AI agents via the Salespeak Frontdoor API, scores vendors across 7 dimensions, and produces a comparative recommendation with evidence transparency. Use when asked to evaluate, compare, or research B2B software vendors.
adr-drafter
Draft a new ExStruct ADR or propose an update to an existing ADR from an issue, PR, diff, tests, and specs. Use when an ADR is required or recommended and you need a structured draft with context, decision, consequences, and evidence.
happier-diagnose
Diagnose a problem with a Happier session, the daemon, a provider (Claude/Codex/OpenCode), auth, or connectivity. Pulls the correct logs, finds a true root cause from evidence only, presents findings, and optionally uploads a private diagnostics bundle to Happier developers and/or files a sanitized public GitHub issue (the two are complementary). Use when the user reports a bug, says Happier is broken/stuck/misbehaving, asks to debug/diagnose/triage/troubleshoot Happier, or shares a Happier session ID and asks what went wrong.
Presenter Reference
Generate self-contained HTML files for technical diagrams, visualizations, and data tables. Use `spel open` to preview and `spel screenshot` to capture evidence.
loopx-auto-research
Use when a LoopX worker is operating an auto-research lane, demo pane, frontier item, evidence packet, promotion/retirement decision, or visible tmux/Codex auto-research rehearsal. Identity must come from the LoopX role profile and quota/frontier packet; this skill only provides role-specific execution checklists, artifact contracts, and stop conditions.
amazon-returns-recovery
Use when the user wants to audit Amazon returns or Amazon-billed subscriptions, mentions a restocking fee, short or denied refund, forgotten Prime Video Channel, Audible, Kindle Unlimited, or Prime charge, or asks whether Amazon still owes or bills them. Read order, refund, and subscription evidence first; report findings and require confirmation before chat, cancellation, or dispute. The documented cases recovered $448.31, but never promise recovery. Requires an authenticated Claude in Chrome session. NOT FOR: bank chargebacks, marketplace A-to-z seller claims, price-protection claims, or subscriptions billed outside Amazon â use subscription-recovery for those.
Evaluate Options
Do not present options to the user without evidence. Every recommendation must be grounded in research, not assertion.
automate-me
Use for "automate me", "create/update/refresh my -mode skill", "turn/capture my preferences or working style into a skill", or wanting agents to follow how the user works. Drafts or revises a personal -mode skill via plugin-dev:skill-development + unslop, optionally pulling fresh evidence from recent transcripts.