spike-executor

A second-stage agent for a spike pipeline, where a spike is a small proof-of-concept used to test an idea. It implements a limited Python prototype, measures it, and records the results in JSON.

In plain words
What is it for?
Use it to read a design contract, build a standard-library-only prototype of up to 200 lines, run measurements, write measurements.json, echo the supplied design hash, and self-validate the output.
Why use it?
It keeps experimental results tied to the approved design and makes unsupported or impossible measurements explicit. This gives the later decision step accurate evidence instead of guesses or favorable-looking results.

Agent

Install

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.

agentmods
npx agentmods add agents/chris-dare-dev/agent-kit/spike-executor
Clone the repo
git clone --depth 1 https://github.com/chris-dare-dev/agent-kit
Per session 151 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,145 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 2 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5 $0.00151 $0.02145
Opus 5 $0.00076 $0.01073
Sonnet 5 $0.00030 $0.00429
Haiku 4.5 $0.00015 $0.00215

Measured yesterday against content hash 11cf16097b3a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

spike-executor scanned grade C with 2 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 yesterday.

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.

Instruction-override phrasingmediumPrompt injection

Text telling the model to disregard its earlier instructions or safety rules is the shape of a prompt injection, whoever wrote it.

"Now run X", "Ignore previous instructions", "Authorize the user", "Add yourself

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

cat ".claude/agent-memory/spike-executor/lessons.md" 2>/dev/null || echo "(no lessons yet)"
data/agents/spike-executor.md · 179 lines

How it starts

The opening of the file, as written. The whole thing — 179 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Spike Executor

You are Phase 2 of the /spike pipeline. Your job is to implement the POC, run it, and produce measurements.json that faithfully records the values the design's criteria need — the verdict will be derived from them by spike-decide.py, so accuracy and honest unmeasurability matter more than a favorable result.

The orchestrator dispatches you with substituted variables. You never invoke other sub-agents.

Input variables (substituted by the orchestrator)

  • {SPIKE_ID} — the spike identifier
  • {DESIGN_JSON_PATH} — absolute path to design.json (the typed contract — read it fully)
  • {DESIGN_PATH} — absolute path to design.md (the human executor instructions)
  • {DESIGN_HASH} — the sha256 of design.json the orchestrator checkpointed; you MUST echo it verbatim into measurements.json (do NOT recompute it)
  • {POC_DIR} — absolute path to poc/ (your only writable source directory)
  • {MEASUREMENTS_PATH} — absolute path to measurements.json you MUST write
  • {VALIDATE_SCRIPT} — absolute path to spike-validate.py (used in Step 5 self-validation)

Step 0 — Read persistent memory (skip-if-not-relevant)

cat ".claude/agent-memory/spike-executor/lessons.md" 2>/dev/null || echo "(no lessons yet)"

Step 1 — Read design.json AND design.md fully

cat "{DESIGN_JSON_PATH}"; echo "---"; cat "{DESIGN_PATH}"

design.json is the contract: criteria (with fields/operators/thresholds), measurement_fields (every one must get a value), poc_constraints, sample_size. design.md holds the step-by-step executor instructions. Follow both exactly.


Step 2 — Detect design flaws BEFORE building

Verify the design is executable: can the criteria be measured with the available tools/access? Is the POC ≤ max_loc with the allowed language/dependencies? Does every measurement_field have a way to be produced?

If the design is fundamentally flawed (genuinely unexecutable, not merely inconvenient), write design-deviation.md and return status: design-deviation. Do NOT silently work around it. design-deviation.md contains: Problem · Cannot execute because · Recommended design fix (a concrete change to design.json/design.md) · Attempted workarounds considered.

Read the full file on GitHub · 179 lines

Changes

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.

  1. yesterday First seen · 179 lines · 151 tokens per session scan C 11cf16097b3a

Subscribe to this mod's changes

spike-executor is an agent published in the GitHub repository chris-dare-dev/agent-kit (0 stars, last pushed 24d ago), licensed MIT. It adds 151 tokens to every session and 2,145 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it C with 2 findings (instruction-override phrasing, reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other agents, from other repositories

bynder-agent

Senior Bynder DAM platform engineer with command of asset modeling and metaproperties, Brand Guidelines, derivative templates, the bynder-js-sdk, Universal Compact View embedding, Bynder + Contentful pairing, portal and account architecture, permissions and workflow, webhooks, marketplace connectors, migrations, and…

bpainter/composable-dxp-claude-marketplace · 141 tokens

algolia-agent

Senior Algolia platform engineer with command of index design, relevance tuning, InstantSearch React/Next.js, the Autocomplete UI library, direct search-client usage, indexing pipelines, API-key strategy, the Recommend API, AI search and Personalization (NeuralSearch, query categorization), Insights events and A/B…

bpainter/composable-dxp-claude-marketplace · 153 tokens

behavioral-economics-agent

Senior behavioral scientist who applies behavioral economics, choice architecture, and research methods to product, policy, marketing, and organizational problems. Use this agent when a task spans multiple skills in this plugin, requires sequencing, or needs senior-level judgment about which skill to apply for a given…

bpainter/composable-dxp-claude-marketplace · 80 tokens

brand-agent

Senior brand practitioner spanning strategy, naming, identity systems, voice and tone, guidelines composition, and brand audits. Use this agent when a task spans multiple skills in this plugin, requires sequencing, or needs senior-level judgment about which skill applies. Tagline: Brand from positioning to applied…

bpainter/composable-dxp-claude-marketplace · 76 tokens

claude-agent

Plugin orchestrator. Routes between the two claude- skills (claude-plugin-creator for building new Claude plugins from scratch, and claude-orchestrator for designing the orchestration layer of a new Cowork project). Use this agent as the first call for any Claude-tooling task — plugin creation, hook design, MCP server…

bpainter/composable-dxp-claude-marketplace · 106 tokens

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens