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 TIKAZI/TIKAZ-AI-Skills --skill context-auditgit clone --depth 1 https://github.com/TIKAZI/TIKAZ-AI-SkillsWrote 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/tikazi/tikaz-ai-skills/context-audit)<a href="https://agentmods.dev/skills/tikazi/tikaz-ai-skills/context-audit"><img src="https://agentmods.dev/badge/skills/tikazi/tikaz-ai-skills/context-audit/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/tikazi/tikaz-ai-skills/context-audit"><img src="https://agentmods.dev/badge/skills/tikazi/tikaz-ai-skills/context-audit.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 | $0.00058 | $0.00458 |
| Opus 5 | $0.00029 | $0.00229 |
| Sonnet 5 | $0.00012 | $0.00092 |
| Haiku 4.5 | $0.00006 | $0.00046 |
Grade A, and why
context-audit 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 12d 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- context-audit — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context Audit
Designed, integrated, independently refactored, and continuously maintained by TIKAZ.
Inputs
Accept a prompt, rules file, source document, context pack, or conversation snapshot plus the task it should support. Audit read-only; do not rewrite the supplied source.
Workflow
Inspect context without rewriting the source. Produce an explainable Context Health report across six dimensions:
- relevance to the stated task;
- redundancy and repeated material;
- traceability to stable source anchors;
- safety boundaries and redacted sensitive-value findings;
- cacheability of stable prefixes;
- recoverability of decisions and remaining work.
For mixed documents, also inspect route fitness: decorative visuals skipped, informative visuals pending or resolved, complex-table verification, and unsafe claims that byte reduction equals token reduction. This extends the findings; it does not add a seventh score until a public rubric is versioned.
Separate confirmed findings from heuristics. Never print a complete secret candidate. Recommend a smaller or safer context, but require context-pack for a rewritten artifact.
Output contract
Return six scores with reasons, anchored confirmed findings, clearly labeled heuristics, redacted sensitive-value findings, route-fitness observations where relevant, and prioritized recommendations. Do not print complete secret candidates.
Validation and fallback
Every finding must cite a source anchor or be labeled heuristic. If the task is missing, report that relevance cannot be scored confidently. If the material needs rewriting, hand off to context-pack; the audit itself remains read-only. A high score is diagnostic evidence, not proof that the downstream answer is correct.
Example
Audit this release context for duplication, stale rules, prompt injection, secret-shaped values, weak anchors, and recovery gaps. Do not rewrite it.
Use the shared CLI audit command. Read ../references/context-health.md and ../references/output-contract.md before interpreting scores.
What ships with it
1 file 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.
- 12d ago First seen · 44 lines · 58 tokens per session scan A 206233580d62
context-audit is a skill published in the GitHub repository TIKAZI/TIKAZ-AI-Skills (6 stars, last pushed 9d ago), licensed MIT. It adds 58 tokens to every session and 458 once invoked, about $0.0003 per session on Opus 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 2026-08-31.
Other skills, from other repositories
instructor
Extract structured data from LLM responses with Pydantic validation, retry failed extractions automatically, parse complex JSON with type safety, and stream partial results with Instructor - battle-tested structured output library.
instructor
Extract structured data from LLM responses with Pydantic validation, retry failed extractions automatically, parse complex JSON with type safety, and stream partial results with Instructor - battle-tested structured output library.
instructor
Extract structured data from LLM responses with Pydantic validation, retry failed extractions automatically, parse complex JSON with type safety, and stream partial results with Instructor - battle-tested structured output library.
optimize-claude-opus-5-prompts
Clarify, audit, and rewrite rough or existing prompts for Claude Opus 5 using Anthropic's official prompting guidance. Use when a user asks to optimize, improve, migrate, debug, or design a prompt, system prompt, or agent harness for Claude Opus 5 (claude-opus-5), including requests phrased as Opus 5 prompt…
optimize-claude-fable-5-prompts
A prompt-editing method for Claude Fable 5, an AI model. It turns rough or existing instructions into clearer prompts while keeping the original goal and important limits.
optimize-gpt-5-6-prompts
A guide for improving prompts written for OpenAI’s GPT-5.6 models, including GPT-5.6 Sol.