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 rtazima/claude-proj-blueprint --skill persistencegit clone --depth 1 https://github.com/rtazima/claude-proj-blueprintWrote 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/rtazima/claude-proj-blueprint/persistence)<a href="https://agentmods.dev/skills/rtazima/claude-proj-blueprint/persistence"><img src="https://agentmods.dev/badge/skills/rtazima/claude-proj-blueprint/persistence.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 5.1 | $0.00059 | $0.01024 |
| Opus 5 | $0.00030 | $0.00512 |
| Sonnet 5 | $0.00012 | $0.00205 |
| Haiku 4.5 | $0.00006 | $0.00102 |
Grade A, and why
persistence 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 7d 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 — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Persistence Mode (Ralph)
Iterative implementation loop that does NOT stop until all acceptance criteria from the PRD pass — or max iterations are reached. Named after Sisyphus: the boulder never stops.
Rules
- Never ask for confirmation between iterations — just keep going
- Never claim "done" without evidence — run the actual test/build commands
- Track iteration count — stop at max iterations even if not all criteria pass
- Report status after each iteration — which criteria pass, which fail, what you'll try next
- Each iteration must make progress — if the same criterion fails 3 times with the same approach, try a different strategy
- Preserve working code — never break passing criteria to fix a failing one
Configuration
- Max iterations: [SPEC] (default: 5)
- Test command: [SPEC] (from CLAUDE.md Commands section)
- Build command: [SPEC] (from CLAUDE.md Commands section)
Workflow
Phase 1: Extract criteria
- Read the PRD at the given path
- Extract ALL acceptance criteria into a numbered checklist
- Print the checklist:
[ ] Criterion 1,[ ] Criterion 2, etc. - Identify the test/build command from CLAUDE.md or project config
Phase 2: Implement (loop)
For each iteration (1 to max_iterations):
── Iteration {N}/{max} ──────────────────────
- Assess: which criteria are NOT yet passing?
- Plan: what specific changes will address the next failing criterion?
- Implement: make the changes
- Verify: run tests/build to check ALL criteria (not just the one you worked on)
- Report:
✅ Criterion 1: passing ✅ Criterion 2: passing ❌ Criterion 3: failing — [reason] ── {passed}/{total} criteria passing ── - If ALL criteria pass → go to Phase 3
- If not all pass → continue to next iteration
Phase 3: Finalize
- Run full test suite one final time
- Run slop-cleaner skill on changed files (if available)
- Print final status:
── Ralph Complete ────────────────────── Iterations: {N}/{max} Criteria: {passed}/{total} passing Files changed: [list] ──────────────────────────────────────── - If max iterations reached with failures, clearly state which criteria still fail
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.
- 7d ago First seen · 99 lines · 59 tokens per session scan A fba2726af485
persistence is a skill published in the GitHub repository rtazima/claude-proj-blueprint (20 stars, last pushed 3mo ago), licensed MIT. It adds 59 tokens to every session and 1,024 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 2026-08-30.
Other skills, from other repositories
truecourse-hooks
Install, configure, or remove the TrueCourse pre-commit hook.
truecourse-analyze
Run TrueCourse architecture analysis on this repository.
init-workspace-documentation
Skill "init-workspace-documentation" from griddynamics/rosetta, covering agent memory.md, agent memory, preventive rules, what worked and what failed.
truecourse-fix
Fix TrueCourse violations that have suggested fixes.
truecourse-list
List TrueCourse violations found in this repository.
release
Guide through the Cortex TMS release process — version bump, changelog, sync, tag, publish. Every step requires user approval.