Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add tomzx/agents --skill identify-codebase-improvementsgit clone --depth 1 https://github.com/tomzx/agentsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/tomzx/agents/identify-codebase-improvements)<a href="https://agentmods.dev/skills/tomzx/agents/identify-codebase-improvements"><img src="https://agentmods.dev/badge/skills/tomzx/agents/identify-codebase-improvements/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/tomzx/agents/identify-codebase-improvements"><img src="https://agentmods.dev/badge/skills/tomzx/agents/identify-codebase-improvements.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00134 | $0.03994 |
| Opus 5.5 | $0.00054 | $0.01598 |
| Sonnet 5.5 | $0.00027 | $0.00799 |
| Haiku 4.5 | $0.00013 | $0.00399 |
Grade A, and why
identify-codebase-improvements scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 22d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 317 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TODAY=!date +%Y-%m-%d REPO=!git remote get-url origin 2>/dev/null | sed -E 's#.*[:/]([^/]+/[^/]+)(\.git)?$#\1#'
Identify Codebase Improvements
Reads an existing codebase as evidence and proposes concrete improvements that are worth doing and are not already tracked in the issue tracker.
Every recommendation must pass two gates: high confidence (the agent read the code and can cite it) and not already an existing GitHub issue.
Dedup is the goal: a pass that re-surfaces known work is noise, so the skill loads the existing issue history with ghx up front and re-checks each candidate against it.
This is a read-only discovery skill. It does not modify code. It produces a ranked recommendation report, and optionally files issues for the recommendations that survive dedup.
The Core Problem This Solves
The issue tracker already holds the known work. A generic audit re-reports it anyway (the same swallowed exception, the same complex function, the same untyped module), which teaches everyone to ignore the audit. This skill inverts the default: it treats every existing issue as a known recommendation, excludes those, and spends its budget finding the improvements nobody has written down yet. It is a loop, not a one-shot list: when a candidate turns out to be already tracked, it drops it and keeps looking, until the code surface stops yielding untracked, high-confidence candidates.
How This Differs From Related Skills
| Skill | Direction | Issue-aware | Output |
|---|---|---|---|
identify-feature-opportunities |
New value to build (what the product could do) | partially | Ranked feature opportunities |
identify-codebase-improvements |
Quality of what exists (what to fix or improve) | yes, ghx dedup is mandatory | Ranked, untracked improvements |
audit-sdlc / audit-* / find-* |
Quality scanners, report findings | no | Findings report |
improve-codebase |
Applies the safe fixes automatically | reads recent PRs only | A PR with verified changes |
check-issues-status |
Finds issues already done (to close) | yes | Close candidates |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 22d ago Changed · +21 lines b00e933dd458
- 25d ago Changed 43b73f875590
- 27d ago First seen · 296 lines · 134 tokens per session scan A a6acda16efb7
identify-codebase-improvements is a skill published in the GitHub repository tomzx/agents (13 stars, last pushed yesterday), licensed MIT. It adds 134 tokens to every session and 3,994 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-03.
Other skills, from other repositories
autoreview
Structured code review through the shared OpenClaw agent-skills installation.
omh-code-review
This is an OMH code-review workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
explore-unknowns
Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when a request is ambiguous or underspecified, the…
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…
asplos-writing-style
Use when drafting or revising ASPLOS prose — engineering the first two pages that rapid review actually reads, stating the cross-layer insight as one quotable sentence, structuring per-layer mechanism sections, writing bounded claims, and fitting everything into 11 self-contained pages that include figures and…