Choosing recipes — the catalog, by function
This guide is curation, not mechanics — it groups the tool-providing recipes by what they actually do, flags where two recipes solve the same problem (so you don’t stack redundant or conflicting tools into one profile), and suggests starting combinations. For the stack manifest schema and build/run lifecycle, see the stacks guide.
What ships and what is planned
The rule, so this list cannot rot silently: shipped means a recipe.yaml under
catalog/recipes/ on main; planned means it is still in catalog/recipes.backlog/. That
directory listing is the authority. harnessed list reports stacks and instances, not recipes, so
it cannot answer this question. First-party recipes (default, mikes-universal-setup,
gh-issue-tracker) and the tracer slices used to exercise the assembly pipeline (greet, ping,
time, floating-recipe) are out of this list’s scope by design.
Shipped (as of 2026-09-17)
- gstack: Garry Tan’s skill suite (browser automation, design, PDF, …), installed via its upstream
./setup· skills recipe - serena: semantic code intelligence MCP (LSP-backed retrieval/editing, 40+ languages) · MCP recipe
- codebase-memory-mcp: codebase knowledge graph (158 languages, C binary) · MCP recipe
- context-mode: context-window optimization / tool-output sandbox · MCP + hooks recipe
- tokensave: pre-indexed semantic code knowledge graph (80+ MCP tools, Rust) · MCP recipe
- agentmemory: persistent memory server (53 MCP tools, HTTP :3111) · service + recipe
- openwiki: agent-written repository wiki with grounded, drift-checkable Claims · MCP + skill recipe
- repowise: repository analysis / retrieval · MCP recipe
- rtk: Rust Token Killer, a CLI proxy compressing dev-command output · CLI + hooks recipe
- superpowers: software-development methodology skill suite (TDD, code review, subagent-driven dev) · skills recipe
- gsd-core: GSD methodology skills · skills recipe
- solidspec: multi-methodology spec-driven development CLI · skills recipe
- caveman: concise-output / token-compression skill · skills recipe
- ccstatusline: status-line renderer the harness invokes for
statusLine· CLI recipe - pulumi: the Pulumi CLI plus its egress allowlist, as pure YAML · tools + egress recipe
- OB1 / Open Brain: shared persistent memory + vector search across AI tools · service + recipe, shipped as the
openbrain-exampletemplate: copy it to your overlay catalog and setOB1_URL, since the recipe carries no upstream endpoint of its own
Planned, still under catalog/recipes.backlog/; packages classified in
recipe-stress-test (repos,
install commands, data models, and architecture gaps each one surfaces):
- headroom: context/tool-output compression before it reaches the LLM · MCP recipe
- gbrain: knowledge brain (synthesis, graph traversal, gap analysis) · service + recipe
- hindsight: memory/recall sidecar (multi-container Postgres stack) · existing service
Code intelligence — pick one pairing, not all three
| Recipe | Mechanism | Scope |
|---|---|---|
codebase-memory-mcp | tree-sitter → graph (158 languages) | Read-only structural queries: search_graph, trace_path, query_graph (Cypher), get_architecture, detect_changes. The cross-language map. |
serena | real LSP, per-language servers | Precise symbol operations: retrieval, rename, refactor, find-references. Explicitly disables its own basic file/search tools (--context ide) to avoid overlapping the harness’s. The scalpel. |
tokensave | tree-sitter → libSQL graph (50+ languages) | Same core mechanism as codebase-memory-mcp (near-duplicate there) plus dedicated token-savings accounting (tokensave gain/cost — per-call token deltas, session/lifetime counters, cost-by-model) and graph-based edit primitives (str_replace, ast_grep_rewrite, rename preview) that graze serena’s territory without a live language server. |
Recommended: codebase-memory-mcp + serena (complementary — map + scalpel). Swap in
tokensave instead of codebase-memory-mcp only if you specifically want its token-cost
telemetry; don’t run all three (codebase-memory-mcp and tokensave overlap directly on
mechanism).
Token / output compression
| Recipe | Compresses |
|---|---|
caveman | The agent’s own generated speech (terse “caveman” mode, ~75% fewer output tokens) |
rtk | Output of explicit dev-command wrapping (rtk git status, rtk read file.rs, 60–90% smaller) |
tokensave | Substitutes graph queries for raw file reads/greps, with instrumented savings |
headroom | On-demand compression of large tool outputs/logs/JSON via MCP tools (headroom_compress/retrieve/stats, 60–95% reduction) — candidate for this group; not yet confirmed as a default pick |
These four address different points in the pipeline (agent speech / explicit CLI wrapping / code-graph substitution / on-demand blob compression) — non-overlapping, safe to combine as needed.
Not in this category: context-mode looks like it should be (it’s about “context”), but its
actual skill (catalog/recipes/context-mode/skills/context-mode/SKILL.md) is explicitly “a
tracer, not a full indexing engine” — it just appends notes to a persist-mounted file to prove
cross-launch persistence works. Zero compression mechanism. It belongs in the memory group below.
Shared cross-project memory — pick one, or none
| Recipe | Backend | Note |
|---|---|---|
agentmemory | Local REST store (harnessed service) + stdio MCP shim | Simplest to stand up yourself |
openbrain-example | OB1 / Open Brain over Streamable HTTP, referenced by URL | A template: copy it to your overlay catalog and set OB1_URL |
gbrain (planned) | Long-lived host service, referenced by URL | Needs the gbrain service already deployed |
hindsight (planned) | Host docker-compose stack (AlloyDB Omni) | Heaviest; needs the compose stack already running |
All of these are “one shared brain across instances/projects” — pick the one whose backend you’re actually willing to operate; don’t run more than one.
Task tracking is not a recipe. The gh-issue-tracker recipe ships the rules and scripts for
working a GitHub Issues backlog, and that is what this repo uses.
Dev methodology / workflow — pick one
| Recipe | Delivers |
|---|---|
gstack | Garry Tan’s ~50-skill suite — browser automation, design, PDF, and the planning/review workflow skills (plan-ceo-review, office-hours, qa, review) |
superpowers | 14-skill composable methodology suite (TDD, systematic debugging, brainstorming, planning, subagent-driven dev, code review, git worktrees) |
gsd-core | GSD methodology skills |
solidspec | Spec-driven CLI methodology: spec → plan → tasks → implement → ship |
Each of these has its own opinion about where the agent tracks tasks. Running two of them means two uncoordinated tracking conventions, not a complementary pairing. Pick one per stack.
harnessed self-test / example stacks
Not workflow tools — these exist to exercise harnessed’s own composition mechanics (skill-only /
stdio-MCP / service-sidecar / mount-widening / dual-harness patterns) and are why
catalog/stacks/livecheck and friends exist.
| Recipe | Proves |
|---|---|
greet | Skill-only recipe (no MCP) |
time | stdio MCP server + skill |
ping | Network-native MCP via a service sidecar |
floating-recipe | Mount-widening (--mount-folder) |
openbrain-example | URL-based remote-MCP template, url_env wiring |
gstack | Multi-recipe composition at scale (~50 skills) — also a real methodology suite, see above |
Suggested starting stacks
Stacks are harness-free: the harness is chosen at launch (harnessed container-run claude --stack dev-core), so none of these name one.
# Everyday coding
name: dev-core
recipes: [default, context-mode, codebase-memory-mcp, serena, rtk]
# + one methodology suite (pick one of gstack, superpowers, gsd-core, solidspec)
name: dev-methodology
recipes: [default, context-mode, codebase-memory-mcp, serena, rtk, superpowers]
# Heavier compression focus
name: dev-compressed
recipes: [default, context-mode, codebase-memory-mcp, serena, caveman, rtk]
Something you only want some of the time, like pulumi for deployment work, does not belong in
any of these. Add it for one launch with --recipe pulumi instead; see
getting started.
See also
- Stacks guide — the
stack.yamlschema and build/run lifecycle. - Recipe-authoring guide — author a new recipe.
- ROADMAP.md — where the product is going, at the epic level.