npc-actions

Before you start: read this whole install prompt, tell me in a few lines what this skill does and what installing it involves, and wait for my go-ahead before you do anything else. Prerequisites: none for the install itself: no account, no API key, no paid service, and nothing else to install first. For using the skill afterwards as described, you need a TypeSafe account and API key (environment variable TYPESAFE_API_KEY) for its Jev typed-judgment service, and a game or mod integration that maps action names onto that game's animations, tasks or clothing swaps; SKILL.md itself names no price and no paid plan for TypeSafe, and the fallback without Jev is a second LLM call with a JSON schema through your own LLM. Never put the key in a file. Install "npc-actions" for me. It is a single SKILL.md playbook for making NPCs physically react to a conversation and follow spoken orders: TypeSafe's Jev returns a typed choice and a probability in about 0.1 s, per-action thresholds and a repeat cooldown decide whether the NPC slaps, punches, hugs, kneels and begs, flees, flips off, squares up, or does nothing. Repository: https://github.com/christos-antono/NPCTalk (the skill is the .agents/skills/npc-actions/ folder at the pinned commit: SKILL.md only; the identical .claude/skills/ copy and the root skills/ copy are duplicates, not the source of this listing, and the rest of the NPCTalk repo is not installed) Pinned version: install the code at commit 59635ceb8ee010599c3caf9d9d1ee1aa114af91d exactly. Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Clone https://github.com/christos-antono/NPCTalk into a temporary folder, check out commit 59635ceb8ee010599c3caf9d9d1ee1aa114af91d, and copy the .agents/skills/npc-actions/ folder from that checkout. Then compare every file in the folder with the fingerprints recorded for this listing on 2026-10-19: [FINGERPRINTS] - SKILL.md: SHA-256 bd264453f5e2eaedbf0a5646596a9621fc51959d9f5fbe3c04cf9d4e97de5555 If a file does not match, it changed since the listing was prepared: do not install it yet, tell me what you found and let me decide. 2. Read SKILL.md in full and tell me in plain words: what the skill does when used, which services it connects to, whether it reads or sends any personal information or keys, and anything that seems unusual. Stop on any red flag and tell me before going further. Confirm these points from the files: - The skill folder holds one file only: SKILL.md (3,384 bytes). There is no code, no script, no reference file and no dependency list in the folder: the skill is instructions for picking an NPC action twice per exchange, plus thresholds, cooldown and animation-mapping rules. - Nothing is sent anywhere during installation itself; the only download at install time is the skill folder from GitHub at the pinned commit. The file names no telemetry and no remote service other than TypeSafe Jev at use time, described next. - At use time, and only then, the design sends data to TypeSafe: two Jev questions per exchange, one before the reply, in parallel with the LLM (is the player giving an order? An action picked with probability at least 0.6 is treated as an order: the NPC does it and the LLM rewrites the line knowing the NPC is doing it), and one after the reply, in parallel with the voice (what does their mood call for?). In the GTA V example code that SKILL.md points to (examples/gta-v/mod/Jev.cs, not part of this skill and not installed), each question is a POST to https://api.typesafe.ai/v1/systemone with header "Authorization: Bearer $TYPESAFE_API_KEY", and the body carries the state Jev judges (what the player said and what the NPC replied), the question instructions and the action criteria. That is a disclosure, not a step to run during this install. - The catalog lives outside the skill folder, in examples/gta-v/mod/ActionCatalog.cs (not installed): each action has a name, a description Jev reads, and its own minimum probability (0.45 for slap, punch, facepalm, square_up, threaten_gun, hug and kiss; 0.5 for headbutt, shirt_off for men, pants_off to underwear, bend_over and kneel_beg; 0.6 for shoot and run_away; 0.7 for middle_finger; none is just talk and is most exchanges). The same NPC does not repeat an action within 60 s unless ordered to. Gender rules (men only) and content rules (underwear stays on) live in game code, not in the prompt. - Mapping to a game lives in the game's code (examples/gta-v/mod/NpcActions.cs in the example, not installed): each action name maps to candidates (animation dictionary and clip, a task such as flee or fight, or a clothing swap), every candidate is checked when the mod loads, the action plays while the line is spoken and aimed at the player, and a director or scripted scene can pin an action (do: kneel_beg) to skip the picker. SKILL.md also describes testing with npcvoice-check actions in the GTA example (25 exchanges through Jev, 10 of them orders) and dev-mode keys in game; do not run any of that during this install. - Without Jev, SKILL.md says a second LLM call with a JSON schema works, but it is slower (0.3-1 s) and its confidence is not calibrated, so thresholds are guesswork; whatever picks, keep the shape: a fixed set, a probability, per-action thresholds, orders checked first. 3. When I say "yes, install", install it as a skill: copy the whole npc-actions folder into the agent's skills directory, in a folder named "npc-actions", keeping its file (SKILL.md). Do not use the repo's plugin, marketplace or npx routes: those install the whole NPCTalk plugin or all of its skills, not this single skill alone. 4. Verify with no network calls: frontmatter valid, file in place. 5. Report what was installed, where, and what I still need to do myself (for example: get a TypeSafe API key myself and set TYPESAFE_API_KEY as an environment variable, never in a file, only if and when I want to use Jev; build or provide the game-side action catalog and animation mapping myself; provide any game and any exchange myself; using the skill is a separate step that only starts when I ask for it, and the use-time send to TypeSafe disclosed above happens only then, on exchanges I run). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.