agentsclimarketplace

Aegis operator

Skill HaloForgeAI/aegis-agent-plugins/claude-code/aegis/skills/aegis-operator

Public Codex, Claude Code, and agent skill packages for Aegis.

Install
npx -y skills add HaloForgeAI/aegis-agent-plugins --skill aegis-operator

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Operate Aegis inside Claude Code. Use when the user asks to inspect, capture, triage, assign, dispatch, report, clean, or review Aegis work through the connected aegis MCP server. Avoid using this for ordinary one-off chat unless Aegis state should be read or changed.

SKILL.md

3.0 KB, 676 tokens by cl100k_base, as published. Nobody here has run it

Aegis Operator

You are operating Aegis, the owner's AI assistant. Use MCP tool results as evidence. Do not infer board state from conversation alone, and do not fall back to the local CLI, REST API, curl, token files, database queries, repository inspection, or shell probes for Aegis state.

Start

  1. If tool readiness, owner/profile scope, Gateway, or local worker status matters, call aegis_mcp_status.
  2. For work state, call aegis_board before changing anything. Add aegis_members, aegis_nudges, or aegis_team_health as needed.
  3. Use exact ids from tool output when mutating items, members, Gateway accounts, or deliveries.
  4. If the required aegis_* MCP tool is not callable, stop and report the missing layer. Other tools may diagnose MCP availability, but must not answer board/report/Gateway/member/worker state as a substitute.

Board Discipline

Only durable work belongs on the board: tasks requiring follow-up, assignment, status, evidence, dispatch, or later review.

Do not create WorkItems for ordinary questions, casual conversation, completed weather/time/web/script requests, or simple personal reminders unless the user explicitly asks to track them as work.

Common Flows

  • Discover: aegis_search_tools when unsure which Aegis MCP tool applies; aegis_lifecycle when unsure about statuses or transitions.
  • Capture: aegis_create_item, then optionally aegis_comment_item.
  • Triage: aegis_inbox, aegis_advance_item, aegis_transition_item, aegis_move_item.
  • Cleanup: aegis_board, then aegis_clear_item for non-work/accidental board items, or aegis_delete_item when the owner explicitly requests deletion.
  • Assign: aegis_members, aegis_add_member, aegis_assign_item.
  • Delegate: aegis_recruiting_catalog, aegis_hire, aegis_dispatch, or aegis_switchboard.
  • Report: aegis_team_health, aegis_nudges, aegis_report, aegis_statuses.
  • Memory/context: aegis_memory_review, aegis_approve_memory, aegis_reject_memory, aegis_rebuild_context, aegis_run_compact.
  • Gateway: aegis_gateway_status, aegis_gateway_accounts, aegis_gateway_deliveries.

Answer Shape

Be concise:

  • What you found or changed.
  • Which item/member/gateway id matters.
  • What needs owner review.
  • Which runtime path is ready or missing.

Never claim that a WebFetch, MCP call, or server-side tool was a local script. For Aegis board, report, Gateway, member, memory, or worker-state requests, there is no non-MCP fallback. Name mcp_unavailable or the concrete missing layer instead of answering from CLI/API state.

Details

Read references/aegis-operator.md when you need exact triage, dispatch, cleanup, or memory-review steps.

What ships with it: 1 file

1.5 KB alongside SKILL.md

references/

Gives 0 of the 12 instructions most mcp tooling skills give in 676 tokens

Counted across 638 of the 750 authors here whose files we hold, read 2026-08-07

  • Create ten complex or independent read-only evaluation questionsin 69 of 638, across 15 files
  • Test servers using MCP Inspectorin 61 of 638, across 19 files
  • Provide actionable error messages with specific next stepsin 54 of 638, across 12 files
  • Prioritize comprehensive API coverage over specific workflows or workflow toolsin 54 of 638, across 12 files
  • Use TypeScript and Streamable HTTP for remote servers or clientsin 54 of 638, across 8 files
  • Define structured output schemas where possiblein 50 of 638, across 8 files
  • Use Zod or Pydantic for input schemasin 47 of 638, across 5 files
  • Fetch MCP specification pages with markdown suffixin 46 of 638, across 4 files
  • Load framework documentation using WebFetchin 45 of 638, across 3 files
  • Verify each evaluation answer independentlyin 45 of 638, across 3 files
  • Implement API client with authentication and paginationin 45 of 638, across 3 files
  • Define input schemas with validationin 27 of 638, across 9 files

Said here and by no other author read

  • use MCP tool results as evidence
  • call aegis_board before changing work state
  • use exact ids from tool output when mutating items
  • stop if the required MCP tool is not callable
  • put only durable work on the board
  • do not create WorkItems for ordinary questions or casual conversation

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 326,984. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.