improvement-loop
Use after resolving a bug, failed task, or unexpected agent behavior to improve the pipeline skills, agents, hooks, or scripts that contributed to the problem. Also proactively suggest improvements when recurring patterns or inefficiencies are observed.
81
72%
Does it follow best practices?
Impact
93%
1.25xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./.claude/skills/improvement-loop/SKILL.mdQuality
Discovery
67%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description does well at answering both 'what' and 'when' with explicit trigger clauses, which is its strongest aspect. However, the actions described are somewhat vague ('improve' rather than specific concrete actions), and the trigger terms, while relevant, could be broader to capture more natural user phrasings. The skill's niche as a post-incident improvement tool could be more distinctly articulated to reduce overlap with debugging or CI/CD skills.
Suggestions
Replace vague verbs like 'improve' with specific concrete actions such as 'refactor error handling', 'add retry logic', 'update configuration', or 'document failure modes'.
Add more natural trigger terms users might say, such as 'postmortem', 'root cause', 'fix pipeline', 'error keeps happening', or 'retrospective'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names a domain (improving pipeline skills, agents, hooks, scripts) and some actions (resolve, improve, suggest improvements), but the actions are somewhat vague—'improve' is not a concrete action like 'refactor', 'update configuration', or 'add error handling'. | 2 / 3 |
Completeness | The description clearly answers both 'what' (improve pipeline skills, agents, hooks, or scripts) and 'when' (after resolving a bug, failed task, or unexpected agent behavior; also when recurring patterns or inefficiencies are observed). The 'Use after...' and 'Also proactively suggest...' clauses serve as explicit trigger guidance. | 3 / 3 |
Trigger Term Quality | Includes some relevant terms like 'bug', 'failed task', 'agent behavior', 'pipeline', 'hooks', 'scripts', and 'recurring patterns', but misses common user phrasings like 'fix', 'error', 'debug', 'postmortem', 'retrospective', or 'root cause analysis'. | 2 / 3 |
Distinctiveness Conflict Risk | The scope is somewhat specific to post-incident pipeline improvement, but terms like 'bug', 'failed task', and 'improve scripts' could overlap with debugging skills, CI/CD skills, or general code improvement skills. The niche of retrospective improvement is implied but not sharply delineated. | 2 / 3 |
Total | 9 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-crafted meta-skill with excellent workflow clarity and actionability. The mandatory gate check, five-step cycle, and anti-pattern prevention are strong differentiators. The main weakness is moderate verbosity — the core message is repeated in multiple forms (overview, gate section, red flags, key insight), and the graphviz diagrams consume tokens that simple lists could replace more efficiently.
Suggestions
Remove the graphviz dot diagrams and replace with compact bullet-point decision trees to save significant tokens while preserving clarity.
Consolidate the 'Red Flags' section into the 'Preventing Improvement Drift' table since they largely overlap, and remove the 'Key Insight' quote block which restates the overview.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is well-structured but verbose for its purpose. The graphviz diagrams add visual clarity but consume significant tokens for decision trees that could be simple bullet lists. Several sections repeat the same core message ('fix first, improve last') multiple times, and the batching/red flags sections overlap with earlier content. However, the tables are efficient and most content is non-obvious guidance. | 2 / 3 |
Actionability | The skill provides highly concrete, actionable guidance: specific commit message formats, exact verification steps per change type, a routing table for which tool to use, classification tables with examples, and template text for user communication. The five-step cycle is immediately executable with clear deliverables at each step. | 3 / 3 |
Workflow Clarity | The workflow is exceptionally clear with an explicit mandatory gate check before improvement work begins, a well-sequenced five-step cycle, verification steps matched to change types, and explicit feedback loops (fix → verify → re-validate). The anti-pattern prevention section adds guardrails against common failure modes in the workflow. | 3 / 3 |
Progressive Disclosure | The skill references other skills (writing-skills, writing-agents) and agents (cc-orchestration-writer, bash-script-craftsman) appropriately, but all content is inline in a single file that runs quite long. The routing table and classification table could potentially be separate references. However, no bundle files exist, so there's no opportunity to evaluate cross-file navigation. The inline organization with clear headers is reasonable but the file is dense. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- aaddrick/claude-desktop-debian
- Commit
76a5a21
- Reviewed
Table of Contents
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.