Tessl • The management layer for AI agent skills

SkillPublic

soul-audit

Re-run or deepen the agent's identity interview. Drives the SHARED bootstrap answer bank (state/interview.json) via `gbrain bootstrap interview --set/--show/--confirm`, then re-renders the identity files with `gbrain bootstrap render --only <FILE> --force`. One answer store, two surfaces: first-boot bootstrap and this re-run skill write the same bank, so an updated answer flows into SOUL.md, USER.md, ACCESS_POLICY.md, and HEARTBEAT.md without hand-editing rendered files. Every answer is the user's own words — never invented.

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/soul-audit/SKILL.md

Quality

Content

81%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is an unusually actionable, well-guarded workflow: exact commands, per-phase keys, structural validation gates, backups, and failure-recovery loops for a mutating operation. Its main weaknesses are redundancy (the never-invent rule stated four times) and a version-sensitive OpenClaw detail; its single external reference is also missing from the bundle.

Dimension	Reasoning	Score
Conciseness	Mostly efficient, but the verbatim-answers rule is repeated roughly four times (intro, IMPORTANT block, 'When it fails', and Anti-Patterns), the 'one source of truth' rationale re-states the description, and a rotting version pin ('OpenClaw 2026.10.6+') is embedded in prose — matching 'could be tightened' rather than 'minor instances of over-explanation'.	3 / 5
Actionability	The command loop gives copy-paste-ready gbrain commands with per-flag comments, each phase lists exact interview keys and render targets, and the output format includes a concrete example message — fully executable guidance covering the common cases.	5 / 5
Workflow Clarity	A mutating skill with explicit validation checkpoints: the read-back ritual (--show, mirror to user, --confirm <hash>), a structural render gate, automatic backups before overwrite, and an error-recovery loop in 'When it fails' (confirm fails -> re-run --status -> read back -> confirm only with agreement).	5 / 5
Progressive Disclosure	A single, well-sectioned file (Contract, command loop, Phases, Output Format, When it fails, Anti-Patterns) with one clearly-signaled one-level reference, but that reference (../conventions/agent-operator-protocol.md) is not present in the skill bundle, leaving a minor organization gap that keeps it below the 5 anchor.	4 / 5
	Total	17 / 20 Passed

Description

67%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A highly specific, distinctive description that names exact commands and output files, but it omits explicit 'when to use' trigger guidance and natural user phrasings, which the guidelines cap at 3 for completeness and leave trigger term coverage partial. Moving one or two natural trigger phrases from the triggers list into the description itself would close both gaps.

Suggestions

Add an explicit 'Use when...' sentence to the description, e.g. 'Use when the user wants to change their agent's identity, personality, or access settings after initial setup.'

Fold 2-3 natural user phrasings from the triggers list (e.g. 'customize agent', 'change my agent's personality') into the description body so trigger term coverage does not depend on the separate triggers field.

Trim the architectural 'one answer store, two surfaces' explanation to one clause — it inflates length without adding trigger or capability information.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions with exact commands ("gbrain bootstrap interview --set/--show/--confirm", "gbrain bootstrap render --only <FILE> --force") and concrete outputs (SOUL.md, USER.md, ACCESS_POLICY.md, HEARTBEAT.md), matching the comprehensive-coverage anchor rather than the 'minor gaps' level 4.	5 / 5
Completeness	The 'what' is clear (drive the shared answer bank, re-render identity files) but there is no 'Use when...' clause or equivalent explicit trigger guidance in the description — 'Re-run or deepen' only weakly implies when — which caps completeness at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	Some relevant natural phrases ("the agent's identity interview", "Re-run or deepen") but missing common user variations like "personality" or "customize"; the truly natural trigger phrasings live in the separate triggers list rather than the description, so it fits 'some relevant keywords but missing variations' rather than good coverage.	3 / 5
Distinctiveness Conflict Risk	The gbrain-specific commands and named identity files carve a clear niche with minimal conflict risk against any other skill, matching the 5 anchor rather than the 'minor overlap risk' level 4.	5 / 5
	Total	16 / 20 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 1 suspicious	Warning

	Total	14 / 16 Passed

Repository: garrytan/gbrain
Path: skills/soul-audit/SKILL.md
Revision: c3e8fd2

Reviewed: 5 days ago

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.

soul-audit

Table of Contents

Is this your skill?