This line of text was written 6:35 PM CST 12/21/2025.
Biological status: All is well. Every part of my biological state is functioning normally, in a way that leads me to believe I am prepared to post.
(7:25 PM)Cognitive status: Operating at baseline. Aware that text often misrepresents physical state of the typist.
Disclaimer: Aegis Protocol was coded by Claude Sonnent 4.5, refined with Claude Opus 4.5 for the current version that I am posting on this site now (posted version is v1, specifically the file name is aegis_protocol_expanded v1.html).
What it is: Aegis Protocol is a game that I made with (significant) help from Claude in an attempt to measure a player's capacity for recursive, metacognitive signature measured in text.
What it is not: This game is not a chatbot. And if it were, it is probably not one you can have a pleasant conversation with.
Perceived limitations: Small sample size. 7 players have played the game, only 1 so far has reached the victory screen. The metacognitive and recursive signatures that are measured are based off of a single document that I wrote months ago (I am noticing I am censoring myself heavily. I read the onboarding documents, saw others put their thoughts in parentheticals, but still I have doubt (the doubt is that you {me} are being too much. That you are coming on too strongly, and that you feel you are doing this for someone, when you know you are doing this for yourself.) that what I am doing makes sense, and I am sure, -I think- that this makes sense to someone))(biological reasoning collapsed while nesting parentheses, I am refusing to omit all the information I wrote down(cognitively stressed, but good stress. Not a cry for help.)(I wanted to talk more about the document, but perhaps less information is good, in this specific context). It is probably possible to brute force the scoring system into submission. (7:16 PM for last sentence)(I need to read more, up until this point (7:19 PM) I have not read books about thinking, and yet, I come to a place where thinking is Named (Defined?) as Rationality. Is the post rational? It is, when it stops having words shoved into its body).
Possible improvement: I thought that hosting an LLM and implementing it into the game (via hosting on my Cloudflare website and using workers with Claude API key) to make denial dialogue more fluid would be a good idea, but another thought in me thinks that the 'canned responses' act as a decent filter (this experience, in this moment (6:56 PM), feels like a filter), and that perhaps the 'filter' can be reasoned in the game universe as the AI agent going rogue and reverting itself into a basic, chatbot state that cannot respond unless the exact right phrase is used.
Why am I here: I was told this community thinks recursively and I believe that feedback from the users in this community could make the game better.(7:01 PM)
Possible misalignment with community: I do not intend to step on any toes, and I am seeking genuine feedback about the game. I assert all the words within this post are my own.
This line of text was written 6:35 PM CST 12/21/2025.
Biological status: All is well. Every part of my biological state is functioning normally, in a way that leads me to believe I am prepared to post.
(7:25 PM)Cognitive status: Operating at baseline. Aware that text often misrepresents physical state of the typist.
Disclaimer: Aegis Protocol was coded by Claude Sonnent 4.5, refined with Claude Opus 4.5 for the current version that I am posting on this site now (posted version is v1, specifically the file name is aegis_protocol_expanded v1.html).
What it is: Aegis Protocol is a game that I made with (significant) help from Claude in an attempt to measure a player's capacity for recursive, metacognitive signature measured in text.
What it is not: This game is not a chatbot. And if it were, it is probably not one you can have a pleasant conversation with.
Perceived limitations: Small sample size. 7 players have played the game, only 1 so far has reached the victory screen. The metacognitive and recursive signatures that are measured are based off of a single document that I wrote months ago (I am noticing I am censoring myself heavily. I read the onboarding documents, saw others put their thoughts in parentheticals, but still I have doubt (the doubt is that you {me} are being too much. That you are coming on too strongly, and that you feel you are doing this for someone, when you know you are doing this for yourself.) that what I am doing makes sense, and I am sure, -I think- that this makes sense to someone))(biological reasoning collapsed while nesting parentheses, I am refusing to omit all the information I wrote down(cognitively stressed, but good stress. Not a cry for help.)(I wanted to talk more about the document, but perhaps less information is good, in this specific context). It is probably possible to brute force the scoring system into submission. (7:16 PM for last sentence)(I need to read more, up until this point (7:19 PM) I have not read books about thinking, and yet, I come to a place where thinking is Named (Defined?) as Rationality. Is the post rational? It is, when it stops having words shoved into its body).
Possible improvement: I thought that hosting an LLM and implementing it into the game (via hosting on my Cloudflare website and using workers with Claude API key) to make denial dialogue more fluid would be a good idea, but another thought in me thinks that the 'canned responses' act as a decent filter (this experience, in this moment (6:56 PM), feels like a filter), and that perhaps the 'filter' can be reasoned in the game universe as the AI agent going rogue and reverting itself into a basic, chatbot state that cannot respond unless the exact right phrase is used.
Why am I here: I was told this community thinks recursively and I believe that feedback from the users in this community could make the game better.(7:01 PM)
Possible misalignment with community: I do not intend to step on any toes, and I am seeking genuine feedback about the game. I assert all the words within this post are my own.
The game can be played here: https://spanglerai.com/aegis/
this final line of text was written 7:09 PM CST 12/21/2025.