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/iamcxa/kc-claude-plugins/continue-dev-flownpx skills add iamcxa/kc-claude-plugins --skill continue-dev-flowgit clone --depth 1 https://github.com/iamcxa/kc-claude-pluginsWhat 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 | $0.00034 | $0.03327 |
| Opus 5 | $0.00017 | $0.01664 |
| Sonnet 5 | $0.00007 | $0.00665 |
| Haiku 4.5 | $0.00003 | $0.00333 |
Grade A, and why
continue-dev-flow 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 — 259 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Continue Dev Flow
Continue by the selected profile's smallest sufficient route.
Resolve authority before policy
- Locate the workflow README via the nearest repository instructions. Require
the unique ordered
kc-dev-flow-static-local-profilestart/end marker pair, with## Local Profileimmediately after start. Read only its frontmatter and marked block; never infer boundaries from headings or open the full README. - Recheck the worktree, branch, shared-state owner, and remote delivery state.
When this item is delivered through a reviewable delivery artifact, list the
open artifacts before creating or reusing a branch. An open unmerged artifact
that carries work this item builds on is the default base: branch from its
source branch and stack, rather than branching from the trunk or waiting for
it to merge.
delivery-branch-base.mdowns the rule and its exceptions, and applies whoever owns the delivery ceremony. - Read the exact committed work item and selected brief. A Pilot or
Production new admission requires one Development Brief containing the
problem, accepted outcome, complete non-goal list, route-back conditions, and
one canonical
## Acceptance criteriasection with unique ascendingAC-Nbullets. A POC item uses its complete v3 decision, falsifier, budget, and stop condition as the Exploration Brief. Do not read current execution state or revalidate and rewrite an already-admitted item's historical headings. - Classify its optional Planning Receipt before provider access. The receipt is
exactly
source,planning-window, andplanning-outcome:- when all Planning Receipt fields are absent, use the Captain-approved committed brief as planning authority and do not invoke a provider reader or comparator;
- when all Planning Receipt fields are present, run provider reconcile only
for the provider-backed branch below. For a new Pilot or Production
admission, first invoke the repository-local profile loader's explicit
--validate-admissionmode; default continuation never selects that mode; and - otherwise report
planning receipt incompleteand stop before reading execution state or mutating it.
- In the provider-backed branch, follow the exact work item's
sourceto the accepted planning item and invoke the repository-local read-only planning reader. Read the item for the problem, decision, success condition, priority, and human-facing status, and obtain the union of the provider's current Ready set for the recorded planning window/outcome and every currently Ready snapshot source even when its current window or outcome moved. Refuse a truncated provider result. Ifsourceis not a resolvable planning link, reportplanning source unavailableand stop before reading execution state. If the reader or its inputs are unavailable, reportplanning reconcile unavailableand stop at the same boundary. Do not promote the admission snapshot into planning authority or invent, migrate, or rewrite its planning item. - Still in that branch, compare the current Ready set with the committed
execution snapshot selected through the adapter's local grouping. Compare
source identity and membership,
planning-window,planning-outcome, accepted goal, and non-goals; classify each difference as added, removed, changed, or moved. Normalize both sets into ephemeral JSON lists whose items containsource,planning-window,planning-outcome,accepted-goal, and string-listnon-goals. Do not commit or reuse those files. Refuse the snapshot unless every item shares the exact window and outcome read from the engaged item. - Invoke the repository-local read-only engage comparator only in the provider-backed branch.
What ships with it
1 file 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.
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 · 259 lines · 34 tokens per session scan A ca39b3c11899
continue-dev-flow is a skill published in the GitHub repository iamcxa/kc-claude-plugins (3 stars, last pushed 2d ago), licensed MIT. It adds 34 tokens to every session and 3,327 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…