Skip to content

Add Hermes Labs skills: lintlang, rule-audit, csv-quality-gate, hermes-jailbench - #59

Open
roli-lpci wants to merge 1 commit into
github:mainfrom
roli-lpci:add-hermes-labs-skills
Open

roli-lpci wants to merge 1 commit into
github:mainfrom
roli-lpci:add-hermes-labs-skills

Conversation

@roli-lpci

Copy link
Copy Markdown

Adds four Hermes Labs skills, one file per augment under catalog/hermes-labs-ai/, each pinned to the commit that last changed its SKILL.md:

  • lintlang: deterministic local static analysis of agent instructions and tool definitions (zero model calls)
  • rule-audit: offline audit of a named system-prompt file for contradictions and coverage gaps
  • csv-quality-gate: CSV preflight check reporting pass/warn/fail with line-number evidence
  • hermes-jailbench: known-pattern jailbreak regression check for endpoints the user is authorized to test

Checks: URLs resolve publicly at the pinned SHAs; SKILL.md names match; installed the pinned PyPI releases in a clean virtualenv and ran each CLI. ai-catalog.json is not edited by hand (regenerated by the workflow, as in #45). Hermes Blind is already in #45 and is not repeated here. Bounded on purpose: these four are the skills with a published package and a documented first result.

Hermes Labs autonomous system, on behalf of Rolando Bosch (roli-lpci)

Copilot AI balanced review requested due to automatic review settings September 29, 2026 14:14

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copilot review overview

🟢 Approval recommended

All entries follow catalog conventions and reference accessible, correctly named definitions at matching immutable commits.

Review effort: Balanced
Findings: None

What changed in this PR

Adds four Hermes Labs agent skills to the contributor-managed catalog.

Changes:

  • Registers deterministic linting, prompt auditing, and CSV validation skills.
  • Registers an authorized jailbreak regression-testing skill.
  • Pins each definition to the latest commit affecting its SKILL.md.
File Description
catalog/​hermes-labs-ai/​lintlang.json Adds the LintLang skill.
catalog/​hermes-labs-ai/​rule-audit.json Adds the rule-audit skill.
catalog/​hermes-labs-ai/​csv-quality-gate.json Adds the CSV quality gate skill.
catalog/​hermes-labs-ai/​hermes-jailbench.json Adds the jailbreak regression skill.

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants