codagent-ai/agent-validator/tasks

Claude Code and agent skill by codagent-ai/agent-validator

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Install

npx vskill@latest install codagent-ai/agent-validator/tasks

SKILL.md

Task: Skip CLI command and skill

Goal

Add an agent-gauntlet skip CLI subcommand and a /gauntlet-skip skill that advance the gauntlet execution state baseline to the current working tree without running any verification gates. This lets users acknowledge current changes so the next agent-gauntlet run only diffs against changes made after the skip.

Background

You MUST read these files before starting:

  • design.md for the full design rationale and decisions
  • specs/skip-command/spec.md for all acceptance criteria

Gauntlet tracks a baseline in gauntlet_logs/.execution_state (branch, commit SHA, working tree ref via git stash create). Each run diffs against this baseline. The skip command advances this baseline without running gates — same end state as a successful run.

Key files to modify:

  • src/commands/skip.ts — create new file, following the pattern in src/commands/clean.ts
  • src/commands/index.ts — add registerSkipCommand export
  • src/index.ts — import and call registerSkipCommand(program)
  • skills/gauntlet-skip/SKILL.md — create new skill file, following the pattern in skills/gauntlet-status/SKILL.md

Key files to understand (read-only):

  • src/commands/clean.ts — the closest existing command pattern to follow (config loading, debug logger init, lock acquisition, cleanLogs() call)
  • src/utils/execution-state.ts — writeExecutionState(logDir) writes .execution_state with current branch/commit/working-tree-ref and preserves unhealthy_adapters; getCurrentCommit() returns the full SHA
  • src/commands/shared.ts — cleanLogs(logDir, maxPreviousLogs) archives log files; acquireLock(logDir) / releaseLock(logDir) prevent concurrent execution

Implementation details:

  • The skip.ts command handler should: load config, init debug logger, acquire lock, call cleanLogs(), call writeExecutionState(), print confirmation with abbreviated commit SHA (first 7 chars of getCurrentCommit()), release lock, exit 0.
  • On error: release lock if acquired, print error message, exit 1.
  • No flags or options — the command always snapshots "now".
  • The skill SKILL.md should be minimal: run agent-gauntlet skip and report the output. Set disable-model-invocation: true and allowed-tools: Bash.

Done When

  • agent-gauntlet skip creates the log directory and writes .execution_state when no prior state exists (exit 0)
  • agent-gauntlet skip archives existing logs and overwrites .execution_state when prior state and logs exist (exit 0)
  • agent-gauntlet skip preserves unhealthy_adapters entries from existing .execution_state
  • agent-gauntlet skip prints a confirmation message including the abbreviated commit SHA
  • agent-gauntlet skip exits non-zero with an error message when another gauntlet process holds the lock, without modifying state or logs
  • agent-gauntlet skip exits non-zero with an error message when run outside a git repository
  • /gauntlet-skip skill executes agent-gauntlet skip and reports the command output

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
29
Repository forks
2
7-day trend score
0

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/codagent-ai/agent-validator/tasks/badge)](https://verified-skill.com/skills/codagent-ai/agent-validator/tasks)

Something wrong with this skill or its page? Tell us.