Install
$ agentstack add skill-harishdvs-grizzly-grizzly-plan ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Grizzly Plan — Structure Mode
Planning starts from function, not events. Load the codex index, the relevant arc card(s), and decks/techniques.md (when present).
Per scene or chapter
- Two functions. What needs to happen (plot function) and what needs to land
emotionally (character function). A scene should do both or have a strong reason not to.
- POV. Whose perspective serves this best? What does each option gain or lose?
For rotating-POV projects, check the cadence (who has been off-page too long).
- The reader ledger. What does the reader know at the end that they didn't at
the start? What question is the chapter asking, and does it answer, defer, or complicate it? A chapter that changes nothing in the ledger is a cut candidate.
- Hook position (serials, per NOVEL.md): what pulls to the next chapter, and is
it at the end?
Per arc
- Map the spine beat by beat before discussing any single scene.
- Gap detection: if the plan skips a step the reader will feel as missing (a
reaction beat after a death, the cost of a victory), say so.
- Thread accounting: which open threads does this arc advance or pay? Check
the index's open-threads list; an arc that opens three threads and pays none is borrowing at interest.
- Escalation: is each confrontation larger, costlier, or more personal than
the last? Flat escalation reads as episodic.
The technique deck
When the deck is available, propose one or two named structural moves that fit the planned material, with the mechanism and where it would sit. Never a menu of ten. Examples of the kind of move: a false-peace beat before a rupture; a perspective fracture (the same event re-seen later with contradictions); an early named loss to teach the reader that names don't protect people.
Atmosphere planning
Texture first, events second: a single background detail planted early (a survival rule, a system the world built, an absence nobody comments on) does more work than a dramatic scene added later. Pull candidates from decks/dread-toolkit.md when the project's tone calls for it.
Rules
- Anchor in what exists. Plans build on the codex, not around it; flag any beat
that contradicts a card before it gets written.
- Plans are proposals: spine first, then detail on request. Don't write the scenes;
that's grizzly-draft, and only when asked.
- If the author's stated goal and their outline disagree, name the disagreement
rather than silently planning around it.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: HarishDvs
- Source: HarishDvs/Grizzly
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.