Skip to main content
NeVibeCode · repository · 41 items

forge-harness: a set by chrono-meta

The chrono-meta/forge-harness repository on GitHub: 41 cards with a description and an install command. They work in Claude Code, Codex, Cursor — 8 platforms in total.

About the repository

GitHub
chrono-meta/forge-harness
Stars
★ 33
License
MIT
Source updated
2026-11-12
In the catalog since
2026-10-27
qp skill qp links test planning, execution, and regression checks for a website or desktop app. Run the full sequence or ask for one stage. CX Codex +1 ★ 33repo stars qp-run skill qp-run walks test cases from qp-plan against a running app through the session adapter and records a verdict for every step. Useful once the plan exists and someone has to execute it. CX Codex +1 ★ 33repo stars qp-regress skill qp-regress is a Claude Code skill that re-runs QA test suites unchanged, checks screen reach, and tracks status deltas per test case. It runs in the regression testing step. CX Codex +1 ★ 33repo stars qp-plan skill qp-plan — a Claude Code skill that prepares an app for testing: builds a surface inventory, runs a MECE check, and designs TSV test cases with named expected sources. Also works in Codex, Cursor, GitHub Copilot, Gemini CLI, and OpenCode. CX Codex +1 ★ 33repo stars video-ingest skill video-ingest pulls both the spoken transcript and the on-screen content — demos, slides, UI — out of a video into agent context. Picks the engine by task, available tooling, and video length. CX Codex +1 ★ 33repo stars verify-bidirectional skill The verify-bidirectional Claude Code skill rechecks an AI recommendation after a user challenges it. It searches for conflicts with earlier decisions and only changes persistent rules when there is supporting evidence. CX Codex +1 ★ 33repo stars sim-conductor skill sim-conductor is a Claude Code skill that runs user-reaction simulations, audits, and quality reviews. It profiles a target artifact before dispatching parallel agents. CX Codex +1 ★ 33repo stars salience-splitter skill salience-splitter — a skill for Claude Code, Codex, Cursor and other agents that splits an overloaded always-loaded context asset (SKILL.md, CLAUDE.md, memory index) into a lean resident layer and an on-demand layer. CX Codex +1 ★ 33repo stars return-path-gate skill return-path-gate is a Claude Code skill that scans SKILL.md files under plugins/ for fire-and-forget chains. It classifies each caller-callee pair as CLOSED or OPEN. CX Codex +1 ★ 33repo stars public-surface-audit skill public-surface-audit scans tracked Git files for private tokens and reports the file, line, match, and severity before a repository is published. CX Codex +1 ★ 33repo stars prompt-regression skill prompt-regression — a skill for Claude Code, Codex, Cursor and other agents that statically checks harness assets after rule or skill edits. Runs a probe suite and reports PASS/FAIL/SKIP per probe — no live model sessions. CX Codex +1 ★ 33repo stars plugin-recommender skill plugin-recommender — a Claude Code skill that takes a task description, searches plugin ecosystems, and returns ranked recommendations with install commands. It checks internal catalogs, GitHub, and the Codex and Claude Code marketplaces. CX Codex +1 ★ 33repo stars pipeline-conductor skill pipeline-conductor runs project checks in sequence and combines their findings in one report. Use it when you want several verification steps to follow a single gated flow. CX Codex +1 ★ 33repo stars phantom-quench skill phantom-quench flags claims in reports, test cases and design docs that the source files do not back up. Catches phantom numbers and unsupported citations. CX Codex +1 ★ 33repo stars persona-roster-expander skill persona-roster-expander turns a seed list of personas into a tiered, judgment-mapped cast bounded by safety rules and exports the structure to a JSON file. CX Codex +1 ★ 33repo stars meta-prompt-builder skill meta-prompt-builder — a skill for Claude Code and five other agent environments. It takes an orchestration plan and turns it into a prompt for each agent, shaped around Goal / Context / Constraints / Done When. CX Codex +1 ★ 33repo stars memory-hygiene skill memory-hygiene spots memory entries that were true once and have drifted since. Ranks each by age and proposes a re-check or an archive move. CX Codex +1 ★ 33repo stars marketplace-gate skill marketplace-gate scores a repository across five readiness checkpoints before submitting it to a plugin marketplace. CX Codex +1 ★ 33repo stars install-wizard skill install-wizard is a Claude Code skill that checks a new project, finds FH setup gaps, and proposes changes one at a time. It supports a dry run without writing files. CX Codex +1 ★ 33repo stars install-doctor skill The install-doctor Claude Code skill checks a project for conflicts before installing meta-harness plugins. It spots duplicates and potential silent overwrites. CX Codex +1 ★ 33repo stars harvest-loop skill harvest-loop — a Claude Code skill that runs an 8-step pipeline when field sessions end. It pulls patterns, conflicts and discoveries back into the FH ecosystem. CX Codex +1 ★ 33repo stars harness-pr-reviewer skill harness-pr-reviewer — a Claude Code skill that checks a submitted PR against the project's baseline files. It compares the diff with CLAUDE.md, memory files and naming rules, then posts a review comment with a merge recommendation. CX Codex +1 ★ 33repo stars harness-doctor skill harness-doctor is a Claude Code skill running in five other agent environments. It scans .claude/, CLAUDE.md, rules and agents for drift, broken references and unnecessary complexity. CX Codex +1 ★ 33repo stars goal-quench skill goal-quench wraps the /goal runner with a tiered safety ladder: pre-run token budget estimates, mid-run signals, and automated post-run quality checks. CX Codex +1 ★ 33repo stars frontier-digest skill frontier-digest — a skill for Claude Code and Codex that aggregates AI engineering developments across technical platforms and research feeds. Gathers intelligence on recent agent architectures and model updates. CX Codex +1 ★ 33repo stars field-harvest skill field-harvest — a skill for Claude Code and five other agent environments. It scans git history in field projects, picks out patterns worth backporting to the shared harness, and logs session work. CX Codex +1 ★ 33repo stars edit-manifest skill edit-manifest — a Claude Code skill that closes the edit loop in a harness: every change to SKILL.md, rules, or CLAUDE.md gets stored with a predicted effect, and the next session checks the prediction against what actually happened. CX Codex +1 ★ 33repo stars dialogue-harvest skill dialogue-harvest — a Claude Code skill that separates what you actually said in a long AI dialogue from the model's agreement. CX Codex +1 ★ 33repo stars deep-clarify skill deep-clarify is a Claude Code skill that turns an unclear request into a written spec through guided questions. It activates for requests to clarify a direction or organize requirements. CX Codex +1 ★ 33repo stars cross-ecosystem-synergy-detection skill The cross-ecosystem-synergy-detection skill for Claude Code, Codex and other platforms finds useful combinations among installed plugins and skills. It ranks the pairs and can check for new matches after a component is added. CX Codex +1 ★ 33repo stars corpus-grounding-expander skill corpus-grounding-expander is a Claude Code skill that pulls public-domain corpora into one verified store for verbatim-relay grounding. It fetches versions, normalizes keys, and records licenses. CX Codex +1 ★ 33repo stars context-doctor skill context-doctor — a Claude Code skill that spots and fixes token waste inside a session. It checks the context footprint, flags repeated full-file reads, and routes noisy command output through a proxy. CX Codex +1 ★ 33repo stars contention-layer skill A skill for Claude Code and other agent CLIs that turns contradictions between skills or research tracks into new skill candidates. When two skills disagree on the same output, contention-layer reads the disagreement as signal. CX Codex +1 ★ 33repo stars auto-decorrelation skill The auto-decorrelation Claude Code skill brings in a verifier from another model family for consequential changes. It runs when a change affects infrastructure, publishing, or work trusted by other skills. CX Codex +1 ★ 33repo stars asset-placement-gate skill asset-placement-gate tells you whether a proposed skill, plugin or agent deserves a place in the shared collection, should stay project-local, or should not exist at all. CX Codex +1 ★ 33repo stars apex-review skill apex-review reviews technical proposals from executive decision-maker viewpoints and builds an HTML slide deck. Useful for surfacing leadership objections before pitching. CX Codex +1 ★ 33repo stars agent-composer skill The agent-composer skill maps a task to available agents and proposes a dispatch plan, including parallel work for independent tasks. It can ask follow-up questions when the request is unclear. CX Codex +1 ★ 33repo stars token-budget-gate skill token-budget-gate works out roughly how many tokens a multi-step task will burn and returns a gate verdict before the agent starts spending. CX Codex +1 ★ 33repo stars mcp-circuit-breaker skill mcp-circuit-breaker halts cascading MCP tool retry loops and prevents wasted session tokens. The skill proposes fallback actions and resets once the tool recovers. CX Codex +1 ★ 33repo stars deliberation skill deliberation — a Claude Code skill that runs design decisions through three layers: an innovator proposes, a devil's advocate challenges, a mediator synthesizes. CX Codex +1 ★ 33repo stars

1–40 of 41

More in NeVibeCode

Project sponsors

Companies that keep this analytics open