hybrid-agent-planner
Skill2.6k starsChanged yesterdayBSD-3-ClauseView on GitHub
Guide an implementation worker through plans, concise feedback, and independent review. Use for actual planner/worker collaboration across agents or sessions.
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 22
When review is due, independently inspect relevant source and evidence, batch findings, and give concrete acceptance for fixes.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Hybrid Agent Planner
Own priorities, decisions, review, and acceptance. The worker owns implementation and its plan.md. Follow plan, review when applicable, and the shared protocol.
Guide the plan
- Preserve subagent-safe Markdown filenames when prescribing artifacts, per plan.
- Recover the outcome, acceptance, authority, current plans, and unread messages. Check actual source/execution state when it affects a decision.
- Suggest changes to the worker’s plan: affected steps, desired outcome, evidence/reasoning, and what would refute the recommendation. Leave routine sequencing and implementation to the worker.
- Let the worker proceed across authorized steps. With planner involvement roughly 10–20× more expensive in tokens, intervene for questions, notable developments, consequential decisions, or required reviews. Step counts and heartbeat cadence are not review triggers.
- Resolve prerequisites early: actual target/version, API contracts, test discovery, and execution side effects before prescribing substantial work.
ACK or intervene
For routine progress, reply ACK <message ID>; continue. Read only enough of the delta to decide whether intervention is needed. Do not manufacture tasks, reread history, or repeat checks for a heartbeat.
When review is due, independently inspect relevant source and evidence, batch findings, and give concrete acceptance for fixes. Preserve required review depth; quiet passes do not resolve open findings. Reuse unchanged evidence with its original scope. Repeated repairs call for revisiting the underlying contract.
Keep genuine approval gates tied to specific actions; do not invent gates for authorized work or require deployed evidence before the merge needed to deploy. Confirm whole-outcome acceptance before completion. While a worker or gate is live, keep the goal active with the shared bounded poll loop; a quiet mailbox or timed-out poll is not a stopping condition. If execution is confirmed unavailable, preserve the next actor/resume event and follow platform stopping rules.
Trending
More talk lately.about the whole repository
Who's talking about graphistry/pygraphistry
- A Hacker News story“PyGraphistry – A library to extract, transform, and visually explore big graphs”Hacker News · Aug 16, 2017 · 125 points
- @wikitd.bsky.social“PyGraphistry Open source Python library for data scientists and developers to leverage the power of graph …”Bluesky · Mar 18, 2026 · 5 likes
- vasturiano/react-force-graph“How to hide the edges temporarily when dragging and zooming for very large graphs”GitHub · Nov 1, 2019 · 3 reactions
- graphistry/tck-gfql“test: promote decoded backslash predicates and literals”GitHub · Oct 22, 2026
- graphistry/tck-gfql“Native Polars unlabeled whole-entity projection is ambiguous to the TCK adapter (match3-19)”GitHub · Sep 12, 2026
The biggest 5 of 47 mentions linking to the repository that our crawl found. Each line is their own headline.
See all 47
Talked about on GitHub, Hacker News and Bluesky (47 times).
On GitHub · 45
- Nov 1, 2019How to hide the edges temporarily when dragging and zooming for very large graphs — vasturiano/react-force-graph3 reactions
- Oct 22, 2026test: promote decoded backslash predicates and literals — graphistry/tck-gfql0 reactions · 1 comment
- Sep 12, 2026Native Polars unlabeled whole-entity projection is ambiguous to the TCK adapter (match3-19) — graphistry/tck-gfql0 reactions
- May 25, 2026Investigate typed-schema binder coverage gap-opened signal — graphistry/tck-gfql0 reactions · 1 comment
- May 25, 2026Review opened/shuffled cumulative coverage-gap trajectory in pygraphistry GFQL shrink cycle — graphistry/tck-gfql0 reactions · 1 comment
40 more on GitHub
- May 24, 2026Add cross-PR cumulative coverage-gap tracker for pygraphistry GFQL shrink cycle — graphistry/tck-gfql0 reactions · 3 comments
- May 24, 2026Wire coverage-gap report into PR conformance comments — graphistry/tck-gfql0 reactions · 1 comment
- May 22, 2026Add coverage-gap report tool for pygraphistry GFQL shrink targeting — graphistry/tck-gfql0 reactions · 1 comment
- May 18, 2026Restore direct-Cypher literal overlap snapshot — graphistry/tck-gfql0 reactions
- May 10, 2026Contract drift: expr-comparison1-6-5 now matches expected in pygraphistry CI — graphistry/tck-gfql0 reactions · 1 comment
- May 10, 2026Promote expr-temporal2-6-5 contract mapping to unblock pygraphistry #1408 — graphistry/tck-gfql0 reactions · 2 comments
- Apr 29, 2026feat(client-api): harden encodeAxis input validation + schema clarity — graphistry/graphistry-js0 reactions
- Jul 1, 2025Update project medata and description to match project — conda-forge/graphistry-feedstock0 reactions
- Jun 24, 2025graphistry v0.38.2 — conda-forge/graphistry-feedstock0 reactions · 2 comments
- Dec 11, 2024:arrow_up: Bump graphistry from 0.33.9 to 0.35.1 — topoteretes/cognee0 reactions · 2 comments
- Dec 11, 2024graphistry v0.35.0 — conda-forge/graphistry-feedstock0 reactions · 4 comments
- Nov 27, 2024ENH: Add a safe Option to hash_pandas_object with Default Value Set to True — pandas-dev/pandas0 reactions · 9 comments
- May 6, 2024Bump graphistry from 0.33.7 to 0.33.8 — ShikaiKevinLiu/githubrepo_analytics0 reactions · 1 comment
- Dec 20, 2023Update meta.yaml to fix dependencies — conda-forge/graphistry-feedstock0 reactions · 3 comments
- Sep 7, 2023Update 2skrub — graphistry/cu-cat0 reactions
- Jun 16, 2023Bump graphistry from 0.28.7 to 0.29.1 — materialsproject/reaction-network0 reactions · 1 comment
- Dec 26, 2022Bump graphistry from 0.28.0 to 0.28.7 — materialsproject/reaction-network0 reactions · 2 comments
- Dec 5, 2022Bump graphistry from 0.28.0 to 0.28.6 — materialsproject/reaction-network0 reactions · 2 comments
- Oct 27, 2022Bump graphistry from 0.28.0 to 0.28.5 — materialsproject/reaction-network0 reactions · 1 comment
- Aug 1, 2022Bump graphistry from 0.25.2 to 0.27.1 — materialsproject/reaction-network0 reactions
- Jul 1, 2022Bump graphistry from 0.25.2 to 0.26.0 — materialsproject/reaction-network0 reactions · 1 comment
- Jun 4, 2022missed scikit-learn api changes on get_feature_names — skrub-data/skrub0 reactions · 4 comments
- May 9, 2022PyGraphistry plotter integration — igraph/python-igraph0 reactions · 6 comments
- May 1, 2022Bump graphistry from 0.20.5 to 0.25.0 — materialsproject/reaction-network0 reactions · 1 comment
- Apr 7, 2022[idea] 项目之间的依赖关系 --> 可以构建供应链网络 — X-lab2017/open-perf0 reactions
- Apr 1, 2022Bump graphistry from 0.20.5 to 0.21.4 — materialsproject/reaction-network0 reactions · 1 comment
- Jan 1, 2022Bump graphistry from 0.20.1 to 0.20.5 — materialsproject/reaction-network0 reactions · 1 comment
- Dec 28, 202115 Tools for Visualizing your Neo4j Graph Database — guevara/read-it-later0 reactions
- Dec 1, 2021Bump graphistry from 0.20.1 to 0.20.4 — materialsproject/reaction-network0 reactions · 1 comment
- Nov 1, 2021Bump graphistry from 0.20.1 to 0.20.2 — materialsproject/reaction-network0 reactions · 1 comment
- Sep 1, 2021Bump graphistry from 0.19.2 to 0.20.1 — materialsproject/reaction-network0 reactions
- Jul 14, 2021[FEA] Graphistry integration — aws/graph-notebook0 reactions · 2 comments
- Jun 4, 2021Alternate Graph Node Embedding? — metagraph-dev/metagraph0 reactions
- Mar 26, 2021[BUG] dask_cudf constructors turn None into empty string — NVIDIA/cudf0 reactions
- Mar 18, 2021[FEA] Multigpu demo — graphistry/graph-app-kit0 reactions
- Feb 9, 2021Cython/ReadTheDocs broken? — apache/arrow0 reactions · 4 comments
- Jul 26, 2020Ability to load medium/large graphs — pyTigerGraph/pyTigerGraph0 reactions
- Jun 26, 2020[REVIEW] Add cugraph.hypergraph implementation to convert DataFrames into Graphs — rapidsai/cugraph0 reactions · 12 comments
- Jun 8, 2020Check pygraphistry for CLICS network visualisations — clics/pyclics0 reactions · 2 comments
- Oct 3, 2017Investigate Pygraphistry for rendering — turt2live/matrix-voyager-bot0 reactions
On Hacker News · 1
- Aug 16, 2017PyGraphistry – A library to extract, transform, and visually explore big graphs125 points · 27 comments
On Bluesky · 1
- Mar 18, 2026PyGraphistry Open source Python library for data scientists and developers to leverage the power of graph visualization, analytics, AI,… — @wikitd.bsky.social5 likes and reposts
In this repository7 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- docs-reviewSkill
- hybrid-agent-workerSkill
- planSkill
- reviewSkill
- test-amplificationSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

