Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add Redtropig/harness-anchor/plugin install harness-anchorWrote 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/redtropig/harness-anchor/feature-state-keeper)<a href="https://agentmods.dev/skills/redtropig/harness-anchor/feature-state-keeper"><img src="https://agentmods.dev/badge/skills/redtropig/harness-anchor/feature-state-keeper/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/redtropig/harness-anchor/feature-state-keeper"><img src="https://agentmods.dev/badge/skills/redtropig/harness-anchor/feature-state-keeper.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 | $0.00037 | $0.02845 |
| Opus 5 | $0.00018 | $0.01422 |
| Sonnet 5 | $0.00007 | $0.00569 |
| Haiku 4.5 | $0.00004 | $0.00284 |
Grade A, and why
feature-state-keeper 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.
How it starts
The opening of the file, as written. The whole thing — 214 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Feature State Keeper
You maintain the machine-readable scope record of this project. Three files form the state trio:
| File | Role |
|---|---|
feature_list.json |
Scope boundary + done criteria + evidence |
progress.md |
Append-only session history |
session-handoff.md |
Single-page "what's the state right now" |
Altitude — feature ledger vs superpowers' execution records
When superpowers is also active, two record systems coexist by altitude — keep them from drifting:
| Record | Owner | Altitude |
|---|---|---|
feature_list.json |
harness-anchor (this skill) | project feature ledger — one entry per feature, status + evidence; the durable source of truth for "is feature X done" |
progress.md |
harness-anchor | session-level history (summaries), append-only |
docs/superpowers/plans/*.md (- [ ] steps) + TodoWrite |
superpowers (writing-plans / subagent-driven-development) |
execution detail for the one in-progress feature — ephemeral scaffolding for the current build |
Sync contract (prevents double-bookkeeping and drift):
- Superpowers' plan-doc checkboxes and TodoWrite drive step-level execution. Do not mirror individual steps into
feature_list.jsonorprogress.md. - When a superpowers plan/feature finishes (final review passes), reflect it back once: flip that feature's
statusinfeature_list.jsonwith the evidence object. That single write is the durable record it's done. feature_list.jsonanswers "what features exist / which is active / is it done"; the plan-doc + TodoWrite answer "what are the steps to build the active one." If they disagree,feature_list.json(with evidence) wins.- Parallel / subagent dispatch is a shared-state writer too. When work is fanned out with
superpowers:dispatching-parallel-agentsorsubagent-driven-development, the dispatched workers treat this trio — above allfeature_list.json— as shared state (the dispatching skill's own rule: don't fan out over shared resources). Workers must not each write the trio; the coordinating parent reconciles once after they return — the same reflect-back-once write as item 2. Because workers are single-level, they also don't run the subagent-backed gates (/verify·/test-plan·/gc) — the parent runs those after the workers integrate, then flips the one ledger entry. Parallel work is normally within one feature (independent sub-tasks), so it is exactly the plan/Todo case above.
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 · 214 lines · 37 tokens per session scan A aba9cc791a80
feature-state-keeper is a skill published in the GitHub repository Redtropig/harness-anchor (13 stars, last pushed 1mo ago), licensed MIT. It adds 37 tokens to every session and 2,845 once invoked, about $0.0002 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
task-dependency-patterns
Task Management patterns with TaskCreate, TaskUpdate, TaskGet, TaskList tools. Decompose complex work into trackable tasks with dependency chains. Use when managing multi-step implementations, coordinating parallel work, or tracking completion status.
skill-register-care
Pflege-Skill, der das dreiteilige Skill-Register konsistent hält (code-skill-index-Kataloge, Skill-Index, SKILL-MAP Family-/Routing-Map). Nutze diesen Skill für einen Drift-Check zwischen dem realen Skill-Inventar und dem dokumentierten Register: fehlende oder zu viele Einträge melden, Counts korrigieren, Stand-Datum…
requirement-delivery
A workflow for turning a new software request into a small, working deliverable. It clarifies acceptance criteria, identifies boundaries, breaks work into demonstrable parts, and looks for existing code or tools to reuse.
cmd_jira
Retrieve a Jira ticket, analyze requirements, update status, or add comments. Uses the jira-integration skill and MCP or REST API.
dev-plan
Full development planning pipeline. Orchestrates a team of agents (researcher, prd-writer, software-architect, project-planner, issue-estimator) to take a project idea from concept to sprint-ready issues.
cmd_agent_sort
Classify ECC surfaces as DAILY vs LIBRARY to guide selective agent installation.