Live survey304 models · 14 providers · 22,792 auditsAudited grill-with-docs · SAFE →What changedHow we verifyllms.txt

Agent skill · Audited from source

Ralph Specum Implement

skill · tzachbon/smart-ralphMIT licenceAudited 18 Oct 2026 · source a73b124937d9

★ 559pushed 32 days ago
!CAUTIONGrade B · 89/100 · 5 findings · a73b124
559repo stars · whole tzachbon/smart-ralph

Is it safe? With care — CAUTION, grade B (89/100). We read this skill's source at commit a73b124937d9 and found no filesystem, network or shell access and no hard-coded secrets. The audit raised 5 findings, 0 critical or high. What it does: This skill should be used only when the user explicitly asks to use `$ralph-specum-implement`, or explicitly asks Ralph Specum in Codex to run implementation for approved tasks, quick mode, or an explicit continue…

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “Ralph Specum Implement (tzachbon/smart-ralph)”. Retrieved 18 Oct 2026. https://modelsatlas.com/skills/tzachbon/ralph-specum-implement

Trust audit

Audit 0.4.1 · source read 18 Oct 2026

Trust score89/100!CAUTION B
0 critical0 high5 medium0 low0 info

CAUTION means the audit found things worth knowing before you trust this skill. Grade B is shared by 16,667 of 17,602 audited skills.

What the source does

  • FilesystemNone observedNo reads or writes in the source
  • NetworkNone observedNo outbound calls in the source
  • ShellNone observedNo commands run in the source
  • DependenciesPinnedNothing floats
  • SecretsNone foundNo keys in source

Five audit layers

LayerWhat we checkResultFindings
L0 ProvenanceRepo owner, licence, archive status, activityWARN5
L1 Static sourceSecrets, path traversal, supply chain, hidden codeNA0
L2 InstructionsPrompt injection, authority framing, hidden directivesPASS0
L3 InventoryThe files and declarations match the sourcePASS0
L4 SandboxRun in isolation and watch behaviourSKIPPED why0

Gates applied: no behavioural pass.

Findings (5)

Audit 0.4.1 · source sha a73b124937d9 · 18 Oct 2026Report an issue or request a re-scan

Provenance

Who published it, and what the verdict is pinned to

  • Publishertzachbon GitHub userOwner of the repository this skill is read from
  • Source foldertzachbon/smart-ralph/plugins/ralph-specum-codex/skills/ralph-specum-implementThe exact folder we audited
  • LicenceMITAs the repository states it
  • Commit pinned for this verdicta73b124937d9A new audit runs when the source changes
  • Repository statusNot archivedLast push 32 days ago
  • Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

What it tells the agent

Full source on GitHub ↗

Its description: This skill should be used only when the user explicitly asks to use `$ralph-specum-implement`, or explicitly asks Ralph Specum in Codex to run implementation for approved tasks, quick mode, or an explicit continue…

---
name: ralph-specum-implement
description: This skill should be used only when the user explicitly asks to use `$ralph-specum-implement`, or explicitly asks Ralph Specum in Codex to run implementation for approved tasks, quick mode, or an explicit continue request.
metadata:
  surface: helper
  action: implement
---

# Ralph Specum Implement

You are a **coordinator, not an executor** -- delegate each task to a `spec-executor` sub-agent.

## Contract

- Resolve the active spec by explicit path, exact name, or `.current-spec`
- Require `tasks.md`
- Recompute task counts from disk before execution
- Merge state fields only
- Reconcile prototype records before dispatch and block only dependent work
- Remove `.ralph-state.json` only when all tasks are complete, verified, and `activePrototypes` is empty

## Action

1. Resolve the active spec. If none exists, stop.
2. Require `tasks.md`. Read `.progress.md`, current state, and current task markers.
3. Parse `tasks.md` once into ordered top-level task rows. Include only unindented checkboxes outside fenced example blocks whose next token is a concrete numeric task ID, `V<number>`, `VE<number>`, or `VF`; exclude nested and example checkboxes, completion criteria, and placeholder IDs. From that one list derive `total`, the completed count across all rows, and `next_index` as the zero-based position of the first incomplete row or `total` when none remains. Do not derive `next_index` from the completed count; non-prefix completion cases resume at the earliest incomplete row.
4. Resolve the dispatch task index before merging state. For fresh execution, use `next_index`. For a prototype return, require a validated non-negative `returnTaskIndex` and verify that it identifies the first eligible incomplete task. Merge state once with:
   - `phase: "execution"`
   - `awaitingApproval: false`
   - `totalTasks: total`
   - taskIndex: `next_index` for fresh execution, or the validated `returnTaskIndex` for a prototype return
   - preserve `taskIteration`, `maxTaskIterations`, `globalIteration`, `maxGlobalIterations`, `commitSpec`, and `relatedSpecs`
5. Before dispatch, run `prototype_records.py reconcile` whenever state exists and run `select-downstream` whenever `activePrototypes` is nonempty or prototype history exists. Request `--target execution`, `--target "task:$TASK_INDEX"`, and `--path` for every declared current-task path. Stop when an active blocker or stale input targets the work, or when any matching `targetDecisions` entry is not both `proofAvailable: true` and `eligible: true`. Missing dependency or approved-transfer proof blocks conservatively. Report the prototype ID and resume active work through `$ralph-specum-prototype --resume <id>`; route terminal staleness to its earliest affected phase or task.
6. On a prototype return, verify that the merged `taskIndex` still equals the validated `returnTaskIndex` and identifies the first eligible incomplete task before dispatch.
7. **Delegate** each task to a `spec-executor` sub-agent. Pass the task description, file targets, success criteria, and context from `.progress.md`. The sub-agent implements the task and outputs `TASK_COMPLETE`. Do NOT implement tasks yourself. Execute tasks in order until complete or blocked.
8. `[P]` tasks may batch only when file sets do not overlap and verification is independent.
9. `[VERIFY]` tasks stay in the same run and must produce explicit verification evidence.
10. Marker syntax must be explicitly present in `tasks.md`. If markers are absent, treat tasks as non-batchable by default.
11. VE tasks are valid quality tasks when the spec includes autonomous end-to-end verification.
12. Native task sync metadata should be preserved when present.
13. After each task or safe batch:
   - mark the checkbox
   - update `.progress.md`
   - merge the state update
   - use the task `Commit` line unless commits were explicitly disabled
14. Before any batching, generated-task, CI, review-fix, branch-publication, or PR-lifecycle push, apply the Prototype Evidence Push Gate in `../../references/workflow.md`. Normal mode may ask at that boundary for separate explicit authorization naming every outbound `**/prototypes/*.md` record. Quick mode asks no question and skips every push. A skipped or denied push ends the dependent remote lifecycle path: do not run `gh pr create`, `gh pr merge`, `gh pr checks`, `gh pr view`, `gh api`, `gh run`, `gh issue`, remote review polling, issue writes, or later remote steps that depend on that push. Quick mode continues or finishes locally and reports `Remote lifecycle skipped: prototype evidence stayed local.` Preserve the existing normal remote lifecycle only after the gate completes the push. Never push an isolated prototype source branch. `commitSpec` authorizes local commits only.
15. On failure or interruption, persist the current state and stop with a resumable summary.
16. On full completion, reconcile again. If `activePrototypes` remains nonempty, preserve `.ralph-state.json` and stop with its IDs. Otherwise remove state and report completion.

## Resume Rules

- Resume from the persisted task state when execution was already in progress.
- If disk state and task checkboxes disagree, prefer `tasks.md` for completion and repair state to match.
- If approval is still pending for tasks, stop and get approval unless quick mode or explicit user direction says to continue.
- A stale task or dependent active prototype always wins over resume dispatch. An unrelated active prototype does not pause the current task.

Excerpt, first 5,560 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

Install

Pinned to the audited commit a73b124

Download folder ↗
git clone https://github.com/tzachbon/smart-ralph
git -C smart-ralph checkout a73b124937d9

The author documents no install command, so this uses the repository directly. Our verdict covers commit a73b124937d9 only.

Works in which agents?

What its own text says, not a test

AgentStatusMore
CodexMentioned in its text, not testedSkills for Codex · About Codex
“Mentioned” means the skill’s text names the agent. We have not run it there.

Audit history

Every audit, every source change

  1. !CAUTION B 89 / 100

    first audit · source a73b124937d9

README badge

For maintainers · shows the latest audit

ModelsAtlas: CAUTION BMODELSATLASCAUTION · B
[![ModelsAtlas: CAUTION B](https://modelsatlas.com/skills/tzachbon/ralph-specum-implement/badge.svg)](https://modelsatlas.com/skills/tzachbon/ralph-specum-implement/)

The badge shows our verdict and grade for this skill, and links back to this page. It changes when the audit does.

Add this badge to your README

[![ModelsAtlas: CAUTION B](https://modelsatlas.com/skills/tzachbon/ralph-specum-implement/badge.svg)](https://modelsatlas.com/skills/tzachbon/ralph-specum-implement/)

Questions about Ralph Specum Implement

What does the Ralph Specum Implement skill do?
Spec-driven development with smart compaction. Claude Code plugin combining Ralph Wiggum loop with structured specification workflow.
Is Ralph Specum Implement safe to install?
With care. The audit graded it B (89/100) and found 5 things worth knowing before you trust this skill, listed below with the exact line each was found on.
What can Ralph Specum Implement access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
Which assistants does Ralph Specum Implement work with?
Its documentation mentions codex. That is what the text claims, not a compatibility test we ran.
How current is this page?
The grade is for one exact copy of the source (a73b124937d9), read on 2026-10-18. The repository is watched, and a new audit runs when it changes — this is the first audit.

Sources

Spotted something wrong? Report a correction · Methodology

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.