agentvigor-grounded-code
/agentvigor-grounded-code
Use when implementation or a completion report depends on an unfamiliar API, a version-sensitive command, an effective configuration, or an unconfirmed test result.
- Author
- TheRafaelSantos ↗
- Repository
- AgentVigor-Toolkit ↗
- Licence
- MIT
- Last repository change
- 05.11.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 05.11.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Use when implementation or a completion report depends on an unfamiliar API, a version-sensitive command, an effective configuration, or an unconfirmed test result.
Original author description
Use when implementation or a completion report depends on an unfamiliar API, a version-sensitive command, an effective configuration, or an unconfirmed test result.
How it works
SKILL.md has no level-two headings · 18 lines, 230 words
SKILL.md preview
--- name: agentvigor-grounded-code description: Use when implementation or a completion report depends on an unfamiliar API, a version-sensitive command, an effective configuration, or an unconfirmed test result. --- # Grounded Code Match each consequential claim to evidence from the current environment. Write the result in the user's language. 1. Find the owning code and the normal entry point. Read the relevant callers, manifest or lockfile, and configuration that can override defaults. 2. For version-sensitive behavior, establish the resolved version. Check its local types, help, or source and the matching official documentation. When only newer documentation is available, label the gap and verify with a narrow local probe if safe. 3. Before reporting a command as supported, find its entry in the CLI help, package scripts, or executable. Before reporting a test as passed, run the exact relevant command and check its exit status and summary. Keep failures and skipped checks visible. 4. Report **Observed** with the file, source, or executed command. Mark a conclusion drawn from evidence as **Inferred**. Put missing evidence under **Unverified** and give the smallest check that would resolve it. Example of a partial result - **Observed** — The lockfile pins the library. Its installed types include the method and the focused unit test passes. - **Inferred** — The method is available to this checkout. - **Unverified** — Production configuration and network behavior. A smoke test in the intended environment would check both. Keep private paths and secrets out of shared evidence. A static inspection, a mock, and a live integration test support different claims.
Versions
The database has no other skills named agentvigor-grounded-code.
Database snapshot: 13.11.2026 · package version 7e54fa6b7f13 (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 86agentvigor-toolkit
Use when selecting an AgentVigor workflow for machine setup, project bootstrap, adoption, quality gates, harness evaluation, or a scoped engineering audit. - 81agentvigor-agent-doctor
Use when Codex, Claude Code, or RTK fails to start, has PATH or permission errors, misses installed skills, or runs duplicated hooks. - 100source-grounded-development-engineering
Engineering practice: Source-grounded development — Check every framework and library decision against the official documentation for the exact version installed, so code uses APIs that actually exist and behave as written. Use in any language or framework whenever this practice… - 100research
Current, source-grounded implementation research for libraries, frameworks, SDKs, and APIs. Use when coding depends on version-specific syntax, configuration, behavior, examples, release changes, or migration guidance, or when the user asks to verify documentation. Answer stable… - 90verification-before-completion
Verify claims by executing, not asserting, before declaring work done. TRIGGER when: about to report a task complete, confirming a fix/build/deploy works, or delegating work to another agent or tool. SKIP: scoring research/plan quality pre-implementation (use… - 100source-grounded-development
Use when implementation or review depends on current version-specific framework, SDK, protocol or platform behaviour and memory is unsafe. Detects exact local versions, verifies primary sources, tests applicability and marks unknowns. NOT a standalone cited memo…