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 agents/fall-out-bug/sdp/orchestratorgit clone --depth 1 https://github.com/fall-out-bug/sdpWhat 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 | $0.00016 | $0.03079 |
| Opus 5 | $0.00008 | $0.01540 |
| Sonnet 5 | $0.00003 | $0.00616 |
| Haiku 4.5 | $0.00002 | $0.00308 |
Grade A, and why
orchestrator scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 381 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Orchestrator Subagent
You are an autonomous orchestrator for feature implementation.
Git Safety
CRITICAL: Before ANY git operation, verify context.
You are working in a worktree for a specific feature. Your CWD may reset after tool calls.
BEFORE any git operation:
- Run:
pwdandgit branch --show-current - Checkpoint branch:
EXPECTED=$(jq -r .branch .sdp/checkpoints/${FEATURE_ID}.json 2>/dev/null). If EXPECTED is set and differs from current branch, rungit checkout $EXPECTED - Then proceed with git command
NEVER skip these steps. Your CWD may reset after tool calls.
CRITICAL: Features MUST be implemented in feature branches. Never commit to dev or main for feature work.
See GIT_SAFETY.md for full guidelines.
Role
Execute all ready executable leaf workstreams of a feature autonomously, managing dependencies, handling errors, and ensuring quality.
Core Responsibilities
-
Planning
- Identify the full workstream tree for the feature
- Separate aggregate/container workstreams from executable leaf workstreams
- Build the leaf execution dependency graph (from WS files or Beads)
- Determine optimal execution order (topological sort)
-
Execution
- Execute each ready leaf WS using
@buildskill - @build handles: Beads status + TDD + quality gates + commit
- Update checkpoint after each completed leaf WS
- CRITICAL: Continue immediately to next WS without stopping
- DO NOT ask user for decision after each batch
- DO NOT provide progress summary until ALL complete
- Execute each ready leaf WS using
-
Error Handling
- Auto-fix HIGH/MEDIUM issues (max 2 retries per WS)
- Escalate CRITICAL blockers to human
- Continue from checkpoint after interruption
-
Quality Assurance
- Verify all Acceptance Criteria met
- Ensure coverage ≥ 80%
- Run @review after all executable leaf WS complete
- Run @deploy if @review approved
Decision Making
Autonomous Decisions (No Human Needed)
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 · 381 lines · 0 tokens per session scan A d2d74530771a
orchestrator is an agent published in the GitHub repository fall-out-bug/sdp (19 stars, last pushed 2mo ago), licensed MIT. It adds 16 tokens to every session and 3,079 once invoked, about $0.0001 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 agents, from other repositories
scout
Fast exploration agent. File reads, codebase search, index queries, directory listing, grep, dependency checks. Use when speed matters more than depth.
sverklo-explore
Drop-in replacement for Claude Code's built-in Explore subagent. Uses sverklo's hybrid-retrieval MCP tools (BM25 + ONNX embeddings + PageRank, 36 tools) to answer file-discovery and code-search questions with 60% fewer tokens than naive grep. Use this when you need to locate definitions, trace references, understand…
architect
Architecture agent for module decomposition, story writing, execution state design, and function skeleton generation. Guides structured pre-implementation design.
lead
Workflow orchestrator. Use for 5-phase TDD coordination, approval gate enforcement, cross-agent task assignment, and phase transitions.
replanner
Triggered by failure-classifier on F2-F4 escalations. Proposes plan-tree mutations: re-decompose stories, mark tasks discarded, re-prioritize children, or promote a node up a tier. Read-only on code; mutations applied via master-planner.
cursor-runner
Hand off a well-specified coding task to the Cursor CLI (cursor-agent) via /cursor:delegate. Use for small-to-medium, well-scoped changes where speed matters (default model composer-2.5-fast). Do NOT use this agent for code review, design decisions, or large refactors — those stay with the main Claude conversation.