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 agentmods add skills/codealive-ai/ai-driven-development/agents-consiliumnpx skills add CodeAlive-AI/ai-driven-development --skill agents-consiliumgit clone --depth 1 https://github.com/CodeAlive-AI/ai-driven-developmentWhat 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 | $0.00086 | $0.03153 |
| Opus 5 | $0.00043 | $0.01577 |
| Sonnet 5 | $0.00017 | $0.00631 |
| Haiku 4.5 | $0.00009 | $0.00315 |
Grade A, and why
agents-consilium scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 220 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agents Consilium
Resolve the entrypoint below, then use only "$CONSILIUM". Select the mode from the user's intent and load only the linked reference needed for that mode.
Entrypoint resolution
Set CONSILIUM to the consilium executable in the scripts subdirectory of the exact
SKILL.md loaded by skill discovery. Verify it before the first call:
test -x "$CONSILIUM" || { echo "agents-consilium entrypoint not found: $CONSILIUM" >&2; exit 1; }
Resolve this from the loaded skill path, never from the caller's working directory, repository
root, or PATH. Do not execute the entrypoint as a relative path. Keep the shell in the
repository the external agent should inspect or change.
Prompt transport is part of the launch
Every review ask invocation must visibly transport a non-empty prompt in the
same shell command: as the positional argument, through an explicit stdin pipe
or heredoc, or with --prompt-file FILE. Never run a bare review ask command
with only options and assume that prompt text from commentary, planning, or the
surrounding agent context becomes stdin; shell execution has no such implicit
connection. For a generated multiline repository review, prefer an explicit
heredoc or prompt file so the complete problem statement and context seed are
delivered atomically.
Recommended review defaults
Use these unless the user asks for a different tradeoff:
- Independent architecture, design, debugging, or planning opinions:
review ask --progress compactwith all enabled profiles. - Routine file or diff review:
review code --progress compact(basic: security + correctness). - High-risk or release-blocking review:
review code --depth super --progress compact. - Use
specialistsonly for a broader mid-cost review without an LLM judge. Useultraonly when the user explicitly prioritizes maximum coverage over cost and latency. - Repository research:
delegate -a grokfrom the target repository root. Tell the worker whether the task is read-only, keep the default steerable session, and usesteer/waitto continue incomplete work. - The
grokprofile is Grok 4.6 and is the default native Grok Build worker. Use the disabled-by-defaultgrok-fastprofile explicitly for fast context research with Grok 4.5. - Codex Sol is an explicit second opinion for difficult specification verification or optimization planning. Select it with
-a codex; do not add it to routine research or the default review pool.
What ships with it
60 files 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.
- ACP-RESEARCH.md 7.7 KB
- assets/agents-consilium-cover.png 1967 KB
- config.example.json 5.1 KB
- config.json 4.9 KB
- prompts/broad-analyst.txt 3.2 KB
- prompts/broad-lateral.txt 2.7 KB
- prompts/judge.txt 4.4 KB
- prompts/probe-generic.txt 2.5 KB
- prompts/review-framework.txt 3.2 KB
- prompts/review-recap.txt 567 B
- prompts/specialist.txt 4.1 KB
- README.md 9.7 KB
- references/configuration.md 3.3 KB
- references/delegate.md 15 KB
- references/review.md 6.0 KB
- references/runtime-contracts.md 10 KB
- references/testing.md 1.1 KB
- scripts/consilium 7.2 KB
- scripts/lib/artifacts.sh 5.0 KB runs code
- scripts/lib/backend_contract.py 14 KB runs code
- scripts/lib/backend_run.sh 35 KB runs code
- scripts/lib/code_review_validate.py 10 KB runs code
- scripts/lib/common.sh 17 KB runs code
- scripts/lib/config.sh 5.2 KB runs code
- scripts/lib/debug_tape.py 11 KB runs code
- scripts/lib/dedup-findings.py 4.2 KB runs code
- scripts/lib/delegate.sh 11 KB runs code
- scripts/lib/discovery-pass.sh 7.6 KB runs code
- scripts/lib/events.py 13 KB runs code
- scripts/lib/human_id.py 2.7 KB runs code
- scripts/lib/judge-runner.sh 7.7 KB runs code
- scripts/lib/mode_policy.py 7.7 KB runs code
- scripts/lib/normalize_stream.py 34 KB runs code
- scripts/lib/progress.sh 2.6 KB runs code
- scripts/lib/prompt_pipeline.py 15 KB runs code
- scripts/lib/quota.py 7.5 KB runs code
- scripts/lib/redact_stream.py 1.5 KB runs code
- scripts/lib/review_ask.sh 15 KB runs code
- scripts/lib/review_code.sh 20 KB runs code
- scripts/lib/review_super.sh 14 KB runs code
- scripts/lib/review_ultra.sh 16 KB runs code
- scripts/lib/steer/__init__.py 155 B runs code
- scripts/lib/steer/adapters/__init__.py 779 B runs code
- scripts/lib/steer/adapters/base.py 3.5 KB runs code
- scripts/lib/steer/adapters/claude.py 13 KB runs code
- scripts/lib/steer/adapters/codex.py 22 KB runs code
- scripts/lib/steer/adapters/grok.py 47 KB runs code
- scripts/lib/steer/adapters/opencode.py 29 KB runs code
- scripts/lib/steer/config_loader.py 1.3 KB runs code
- scripts/lib/steer/control.py 29 KB runs code
- scripts/lib/steer/jsonrpc.py 8.8 KB runs code
- scripts/lib/steer/mailbox.py 9.7 KB runs code
- scripts/lib/steer/registry.py 12 KB runs code
- scripts/lib/steer/supervisor.py 41 KB runs code
- scripts/lib/steer/util.py 8.6 KB runs code
- scripts/lib/steer/waiter.py 6.5 KB runs code
- scripts/lib/terminal_guard.py 5.6 KB runs code
- scripts/lib/validate_judge_output.py 6.4 KB runs code
- scripts/lib/workflow_plans.py 14 KB runs code
- scripts/lib/workflow_runner.sh 8.3 KB runs code
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.
- 2d ago First seen · 220 lines · 86 tokens per session scan A 567a18a63759
agents-consilium is a skill published in the GitHub repository CodeAlive-AI/ai-driven-development (131 stars, last pushed 4d ago), licensed MIT. It adds 86 tokens to every session and 3,153 once invoked, about $0.0004 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
map-auto
Single-entry autonomous autopilot: routes a task through the existing MAP workflows via routetask, then drives the selected chain (map-plan -> map-efficient -> map-check -> map-review, as routed) end-to-end to a committed feature branch in one session, auto-approving routine workflow-control holds and hard-stopping on…
map-resume
Resume an interrupted MAP workflow from .map/ /stepstate.json checkpoint. Use when returning after context exhaustion, /clear, or a session crash mid-workflow. Do NOT use to start new work; use map-plan or map-efficient.
map-skill-eval
Evaluate a /map- skill's trigger accuracy and cost. Use when asked to measure skill trigger accuracy, run an eval-set, or check token/duration cost via mapify skill-eval. Do NOT use to plan or implement; use map-plan or map-efficient.
create-workflows
Creates and revises portable declarative AI workflow files using Markdown frontmatter, DAG nodes, provider IDs, artifact references, imports, reusable inputs, findings artifacts, and review loops. Use when asked to author, improve, modularize, or validate .task.md workflows for CLI-based multi-agent workflow…
intuitive-doc
Create and maintain an intuitive human documentation surface for AI-agent-developed repos. Use when humans should only need README.md, ARCHITECTURE.md, STATUS.md, and docs/human/ while planning logs, generated docs, retrospectives, ADR detail, and implementation evidence stay in AI-agent-only folders.
skill-runner
Run a bounded development task through named skills in an isolated, tmux-backed Codex or Claude session. Use for durable or artifact-sensitive worker phases, supervised skill runs, and post-run skill evaluation.