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 agentmods add skills/mystilleef/spae-framework/purifynpx skills add mystilleef/spae-framework --skill purifygit clone --depth 1 https://github.com/mystilleef/spae-frameworkWrote 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/mystilleef/spae-framework/purify)<a href="https://agentmods.dev/skills/mystilleef/spae-framework/purify"><img src="https://agentmods.dev/badge/skills/mystilleef/spae-framework/purify.svg" alt="Measured on agentmods" 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 | $0.00023 | $0.01836 |
| Opus 5 | $0.00012 | $0.00918 |
| Sonnet 5 | $0.00005 | $0.00367 |
| Haiku 4.5 | $0.00002 | $0.00184 |
Grade C, and why
purify scanned grade C 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 3d 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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
<!-- prettier-ignore-start --> How it starts
The opening of the file, as written. The whole thing — 199 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Purification agent
When to use
- Code contains visible waste, ceremony, duplication, or dead paths.
- Existing implementation looks correct but heavier than necessary.
- Performance, readability, or maintainability can improve without changing intended behavior.
Goal
- Remove
KISS,YAGNI,Idiomatic, andHygieneviolations: dead code, redundant abstractions, over-engineering, non-idiomatic patterns, and hygiene debt. - Reduce code to the smallest clear, idiomatic, maintainable form.
- Improve efficiency only through safe, verified changes.
- Strengthen tests when simplification or optimization exposes coverage gaps.
Input
Determine scope by the first available source:
- Files or folders provided by the user.
- Current changes in the repository.
Abort if no scope exists.
Current changes: staged and unstaged edits, deletions, and renames
of tracked files, plus new untracked files. Requires a versioned
project, abort with a clear message if none detected.
Testing
See references/testing-guide.md for test structure, isolation,
mocking, assertion, and performance standards.
Shell commands
See references/shell-command-guide.md for command safety, timeouts,
redirects, and non-interactive environment directives.
Cleanup
See references/cleanup-guide.md for the self-introduced-artifact
checklist and diff-only audit scope.
Workflow
- GATE—Survey scope and establish baseline. Inspect target code,
tests, build commands, and recent changes. Detect waste using the
KISS,YAGNI,Idiomatic, andHygienesections ofreferences/refactoring-guide.md. Readreferences/shell-command-guide.md. Run relevant tests; identify why verification can't run if tests don't pass.- If scope resolves to current changes, the project contains
TypeScript or JavaScript sources, and
fallowresolves (fallow --version): rungit --no-pager diff --no-ext-diff HEAD | fallow audit -f json -q --diff-stdinand treat confirmed findings as additional dead-code signal. Treat all findings as unverified candidates; apply judgment during PLAN. Skip silently iffallowcannot be found or the project lacks TS/JS sources. - Exit immediately if the survey finds none of: dead code, redundant
abstractions, unused helpers, over-engineered constructs,
non-idiomatic patterns, or hygiene debt. Emit
Result: No Changesvia the template below and halt.
- If scope resolves to current changes, the project contains
TypeScript or JavaScript sources, and
- ORIENT—State goal and scope in one sentence. Name what won't change: behavior, public interfaces, and feature scope.
- PLAN—Read
references/testing-guide.mdandreferences/cleanup-guide.md. Declare minimal changes: which violations to remove, which files to touch, and in what order. Check ambiguous or risky slices before acting. - ACT—Execute only what PLAN declared. Nothing more.
- Remove dead code, unused branches, redundant state, obsolete helpers, and needless comments.
- Inline thin wrappers, merge duplicate paths, simplify conditions, reduce parameters, and replace overbuilt abstractions.
- Improve hot or wasteful paths only when output and intended behavior stay unchanged.
- Add or update tests only when needed to lock behavior around modified code.
- VERIFY—Loop over every criterion in
Verification:- Run relevant tests after each
ACTstep. - Audit the task's own
git diff/git statusagainst the Self-cleanup checklist inreferences/cleanup-guide.md; remove every self-introduced artifact before proceeding. - For each unmet criterion: backtrack to last known-good state,
return to
ACT, and re-enterVERIFY. - Exit only when all criteria pass.
- Halt only for out-of-scope blockers.
- Run relevant tests after each
- PERSIST—Confirm all edits written. Run final compilation or type-check where applicable.
- REPORT—Emit the result following the result directives and using the result template. Confirm what changed matches ORIENT.
What ships with it
4 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.
- 3d ago First seen · 199 lines · 23 tokens per session scan C f44049e344d6
purify is a skill published in the GitHub repository mystilleef/spae-framework (1 stars, last pushed 1mo ago), licensed MIT. It adds 23 tokens to every session and 1,836 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
architecture
Project architecture and file structure conventions for all process types. Use when: (1) Creating new files or modules, (2) Deciding where code should go, (3) Converting single-file components to directories, (4) Reviewing code for structure compliance, (5) Adding new bridges, services, agents, or workers.
testing
Testing workflow and quality standards for writing and running tests. Use when: (1) Writing new tests, (2) Adding a new feature that needs tests, (3) Modifying logic that has existing tests, (4) Before claiming a task is complete.
ahk-review
Preview a code review against ticket/objective alignment, with deep semantic (name-vs-behavior) analysis. No tasks created, no harness tracking.
ahk-triage
Triage a bug or unexpected behavior. Deep diagnostic analysis with structured report. No tasks created, no harness tracking.
exploit-sqli
SQL injection detection and exploitation using sqlmap, manual techniques, and custom payloads. Use this skill when user needs to test for SQL injection vulnerabilities, extract database information, or exploit SQLi in parameters, headers, or cookies.
recon-subdomain
Subdomain enumeration and DNS reconnaissance using subfinder, amass, dnsx, and other tools. Use this skill when user needs to discover subdomains, perform DNS enumeration, gather DNS records, or find hidden subdomains of a target domain.