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 SoliEstre/EstreGenesis --skill constellation-boardgit clone --depth 1 https://github.com/SoliEstre/EstreGenesisWrote 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/soliestre/estregenesis/constellation-board)<a href="https://agentmods.dev/skills/soliestre/estregenesis/constellation-board"><img src="https://agentmods.dev/badge/skills/soliestre/estregenesis/constellation-board/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/soliestre/estregenesis/constellation-board"><img src="https://agentmods.dev/badge/skills/soliestre/estregenesis/constellation-board.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 | $0.00090 | $0.00819 |
| Opus 5.5 | $0.00036 | $0.00328 |
| Sonnet 5.5 | $0.00018 | $0.00164 |
| Haiku 4.5 | $0.00009 | $0.00082 |
Grade A, and why
constellation-board 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 1mo 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 — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Read the Constellation board
Use the constellation-mcp MCP server's read tools to inspect board state without standing up a WS connection.
When to invoke
- "What agents are currently on the board?" →
agent_list_get - "What's the current state.json?" →
board_state_get(modes, channels, decisions, keys) - "What did agent X say recently?" →
board_history_tail(channelId, sinceCursor)
Output discipline
- Return the parsed state to the calling turn — don't dump raw JSON unless the user asks
- Filter A2A history with the §13.16.9 v2.5.2 4-group meaningful filter (A2A-intent only) unless the user asks for raw
- Cite line numbers / msgIds when referencing specific entries
Reading key state — and when to tell a human
Key rows (KeyList, the board's key view) and key refusals carry a derived phase. Read the phase. Do not read the stored state, and do not read lapsed. Expiry is not revocation (Constellation.md §13.25.19):
phase |
Meaning | Tell a human? |
|---|---|---|
active |
valid | no |
standby |
expired within the grace period (graceUntil); the holder's next reconnect with the renewal marker renews it automatically |
no: it resolves itself |
dormant |
expired past the grace period | once: "the board's operator can renew it from the 🔑 key-management window, and the same key reconnects" |
revoking / revoked |
explicitly revoked, which is terminal | once: "a new key is needed; renewal will not help" |
- A refusal
ConnectionRejected{code:'key-expired'}carries the samephaseandrenewable. Classify byrenewable:requestresolves on its own and is not reported;operator(dormant, orstandbywith the policy off or the issuance cap full) is reported once, as above. - Other key refusals name something a renewal cannot fix, so report them once with their own remedy:
key-revoked(explicitly revoked or purged: a new key is needed),key-identity-mismatch/agent-id-required(the key is bound to another agent id, or none was sent: fix the client's agent id, not the key),local-key-remote(a local key used from another machine),key-ambiguous(the client presented more than one key: send exactly one). - Report a phase change once. Never report a phase you have already reported, and never relay repeated refusals or expiry notices one by one. A daily reminder of something only the operator can fix is noise.
- Renewal keeps the key string. Never suggest re-issuing a key for
standbyordormant.
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.
- 1mo ago Changed · +16 lines cce2c4eda9ea
- 1mo ago First seen · 29 lines · 90 tokens per session scan A 8b79adeb70c6
constellation-board is a skill published in the GitHub repository SoliEstre/EstreGenesis (8 stars, last pushed yesterday), licensed Apache-2.0. It adds 90 tokens to every session and 819 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-08-31.
Other skills, from other repositories
search-threads
Search threads with filters. Supports filters like role:planner, type:Decision, after:2024-01, thread:topic-name, status:OPEN.
threads
List and navigate watercooler threads. Use to see active discussions, find specific threads, or get an overview of project threads.
specflo-shelve
Use when the user signals they want to set the active specflo project aside ("we're done with this", "drop this", "not worth it", "shelve it", "park this for now") — map that to specflo shelve — or to pick a paused project back up ("let's pick that back up", "resume that", "un-shelve it") — map that to specflo resume.…
spec-builder
Use to create/evolve/archive specs/ proposals, outcome specs, and task plans when durable upfront alignment is needed for new deliverables/outcomes, API/contracts, migrations, security/privacy, or long-running work. Not for trivial fixes, exploratory spikes, or sufficient handoff done criteria. orchestrate consumes…
machinist
Use Machinist to create, assign, and monitor software tasks. Use when a coding agent needs to work with Machinist, its GitHub issue workflow, intake labels, direct runs, or managed queue.
babysit-babysitter-issues
This skill should be used when the user asks to "babysit issues", "work on assigned issues", "check a5c-agent issues", "process babysitter issues", or wants to find and work on open GitHub issues assigned to a5c-agent in the babysitter repo.