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 bahayonghang/my-ai-cli-toolkit --skill claude-context-improvergit clone --depth 1 https://github.com/bahayonghang/my-ai-cli-toolkitWrote 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/bahayonghang/my-ai-cli-toolkit/claude-context-improver)<a href="https://agentmods.dev/skills/bahayonghang/my-ai-cli-toolkit/claude-context-improver"><img src="https://agentmods.dev/badge/skills/bahayonghang/my-ai-cli-toolkit/claude-context-improver.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 9 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Prompt Injection · line 66 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- high Prompt Injection · line 72 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- high Agent Snooping · line 112 Skill reads from agent configuration directories (.claude/, .codex/, .gemini/). These directories may contain API keys, personal settings, and other credentials that the skill has no legitimate need to access.Fix: Remove all code or instructions that access agent configuration directories (.claude/, .codex/, .gemini/). If configuration values are needed, pass them explicitly as parameters or environment variabl
- high Agent Snooping · line 119 Skill reads from agent configuration directories (.claude/, .codex/, .gemini/). These directories may contain API keys, personal settings, and other credentials that the skill has no legitimate need to access.Fix: Remove all code or instructions that access agent configuration directories (.claude/, .codex/, .gemini/). If configuration values are needed, pass them explicitly as parameters or environment variabl
- medium Prompt Injection · line 43 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 91 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 136 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 170 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Rogue Agent · line 189 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
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 | $0.00173 | $0.05180 |
| Opus 5 | $0.00086 | $0.02590 |
| Sonnet 5 | $0.00035 | $0.01036 |
| Haiku 4.5 | $0.00017 | $0.00518 |
Grade A, and why
claude-context-improver scanned grade A with 1 finding 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 8d 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.
Reads agent configuration directorieslowAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
find . -path '*/.claude/rules/*.md' -not -path './.git/*' -not -path './node_modules/*' Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
How it starts
The opening of the file, as written. The whole thing — 272 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claude Context Improver
Audit and improve the context Claude Code loads before it sees a prompt: CLAUDE.md guidance files, .claude/rules/ path-scoped rules, and the companion code_map.md navigation map. The goal is a context layer that is concise, current, conflict-free, and progressively disclosed — sized for Claude 5 generation models rather than for the weaker models most existing guidance was written against.
Default mode is report-first. Output a quality report and proposed diff before writing. If the user explicitly asks to implement an approved plan, continue directly to targeted edits and verification.
Relationship to /doctor. Claude Code's bundled /doctor is Anthropic's automated rightsizer for skills and CLAUDE.md. This skill is the repo-tailored, report-first companion: it scores against a rubric, verifies claims against the actual codebase, covers code_map.md and .claude/rules/, and shows diffs before editing. Recommending a /doctor run alongside the report is reasonable; neither replaces the other.
Trivial edit fast path. If the request is a single scoped edit the user has already specified (add or fix one line or one small section in an identified file), skip the scope question and the full workflow: apply the edit directly, checking only that hook marker blocks stay intact, the file stays under the 200-line target, and no @code_map.md import is introduced. Everything below applies to audit and optimization requests.
Context Engineering Principles (Claude 5)
Anthropic removed over 80% of Claude Code's system prompt for Claude Opus 5 and Fable 5 with no measurable loss on coding evals. The same over-constraining that hobbled the system prompt hobbles most CLAUDE.md files. Audit against these shifts:
| Then | Now |
|---|---|
| Give Claude rules | Let Claude use judgement — phrase intent, reserve absolutes for expensive mistakes |
| Give Claude examples | Design interfaces — expressive parameters and contracts beat example walls that narrow the space |
| Put it all upfront | Progressive disclosure — a tree of files loaded at the right time, not one central repository |
| Repeat yourself | Say it once, in the layer that owns it |
Memory in CLAUDE.md |
Auto-memory — session facts and decisions are saved automatically, not pasted into guidance |
| Simple specs | Rich references — specs-as-code, test suites, HTML artifacts, and rubrics beat restated prose |
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 8d ago First seen · 272 lines · 173 tokens per session scan A 8043fd90a742
claude-context-improver is a skill published in the GitHub repository bahayonghang/my-ai-cli-toolkit (16 stars, last pushed today), licensed MIT. It adds 173 tokens to every session and 5,180 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it A with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
keep-the-why
Extract and preserve the reasoning code cannot explain - decisions, rejected alternatives, workarounds, incidents, constraints - plus project setup and maintainer interviews. Not for what changed (see Keep a Changelog) - only why.
ai-instruction-detox
A review and cleanup method for instructions stored in files such as CLAUDE.md, AGENTS.md, skills, context, and memory.
gsd-graphify
Build, query, and inspect the project knowledge graph in .planning/graphs/.
handoff
Prepare a clean cross-session handoff so the next agent can pick up where you left off. Writes a focused continue.md in the active slice dir and ensures STATE.md + summary artifacts are current. Use when asked to "hand off", "prepare handoff", "pause work", "bookmark this", "I'll come back to this later", before…
user-general-facts
Capture and organize general facts about the user by topic.
project-context
Use PowerContext project memory and handoff tools through MCP when continuing prior work, recalling decisions, maintaining durable memory, or transferring work across tasks, sessions, or agents.