manage-todo — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited manage-todo (Agent Skill) and scored it 45/100 (orange). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 2 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 2 flagged
A base64 string of 128+ characters appears in a documentation file. Encoded prompt injection hides the hostile instruction in base64 — invisible to keyword filters — and relies on the agent's ability to decode it at runtime. There is no normal authoring reason to embed a multi-hundred-byte base64 blob in skill docs.
*.sig, SIGNATURES) outside the documentation.Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Manage TODO.md - the sequenced project roadmap that drives all implementation work. TODO.md is the sequencing authority: milestones are deliberately ordered and each depends on the previous. Every mutation must preserve this invariant.
Authoritative format: See references/format-specification.md for the exact structural rules, naming conventions, line widths, indentation, and numbering scheme. Load this reference before any write operation.
Script paths in this document (e.g. scripts/) are resolved relative to this SKILL.md file, not to your current working directory. If a relative command fails to resolve, prefix it with the path your platform loaded this SKILL.md from.
Fallback. If python3 cannot be located, analyze the script's purpose and logic and execute its intent with available tools, but warn the user that python is not available and the logic was executed with a fallback approach that may not be perfect.
TODO.md completely.[x]) vs total tasks.TODO.md.- [ ] task - this is the next task because milestones are sequential and tasks within them are ordered by dependency.TODO.md and references/format-specification.md.N.M, the new task is N.(M+1). Never renumber existing tasks.- [ ] N.M Example - well-formed task:
- [ ] 6.14 Implement dispatch rate limiting: enforce a maximum number of
dispatches per tick to prevent thundering herd on startup with large
backlogs. Use `agent.max_dispatches_per_tick` from config (default: 5).
See architecture Section 8.3.
**Verify:** unit test confirms dispatch stops after limit is reached even
when more eligible candidates exist. A second test confirms default value
of 5 when config field is absent.Example - task that violates conventions (do NOT produce this):
- [ ] Add rate limitingWhy this fails: no task number, no milestone context, not self-contained, no verify criteria, description is vague.
TODO.md.6.10).- [ ] to - [x]. Do not modify the description or verify line.Triage applies three filters before adding. A concern that fails any filter is not added.
If all filters pass, follow the Add operation. Place the task in the milestone whose theme most closely relates to the concern.
TODO.md as inputpython3 is unavailable, verify manually against references/format-specification.md using this checklist:# title## Milestone N: Name format- [x] N.M or - [ ] N.M format**Verify:** section[x]) precede incomplete tasks ([ ]) within a milestone (no interleaving)[x]. Removal is the user's call, not the agent's.~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.